IncobotulinumtoxinA for the treatment of lower-limb spasticity in children and adolescents with cerebral palsy: A phase 3 study¹

2.1Participants

The TIM study recruited ambulant and non-ambulant children and adolescents, aged 2–17 years, with unilateral or bilateral LL spasticity due to CP. To be enrolled, patients were required to have a clinical need determined by clinicians for treatment of either unilateral pes equinus with treatment of ipsilateral flexed knee or adducted thigh, or bilateral pes equinus. In addition, patients had to have a clinical need for incobotulinumtoxinA 16 U/kg treatment of LL spasticity and an Ashworth Scale [32] of the plantar flexor (AS-PF) score of ? 2 (at least unilaterally; bilaterally for the treatment of bilateral pes equinus) at screening and baseline. Investigators evaluated the clinical needs of each patient and selected the appropriate treatment pattern. Patients could be BoNT treatment naïve (i.e., no BoNT treatment within 12 months prior to study treatment) or have previously received treatment with BoNT (i.e., > 16 weeks prior to study treatment).

Key exclusion criteria included fixed contracture or predominant forms of muscle hypertonia other than spasticity (e.g., dystonia) in the target limbs, surgery for pes equinus in the target limbs within 12 months prior to screening or planned within the study period, hip flexion requiring orthopedic management and/or BoNT-A injection or limitation of hip abduction to < 40°, or a prediagnosed migrational percentage > 30. Patients with a severe neurological diagnosis and comorbidity outside the spectrum of CP and those with pure dyskinetic CP or mixed CP with predominantly dyskinetic movements were also excluded.

2.2Study design and treatment

The TIM study was a prospective, double-blind, randomized, multicenter, parallel-group, phase 3 study conducted in 45 sites across 14 countries worldwide. Eligible patients were randomized 1:1:2 to three parallel incobotulinumtoxinA dose groups, respectively: low dose: 4 units/kilogram (U/kg) body weight (BW), maximum total dose 100 U; mid dose: 12 U/kg BW, maximum total dose 300 U; high dose: 16 U/kg BW, maximum total dose 400 U.

Two LL clinical patterns were selected for treatment for each patient, one of which was required to be pes equinus on one side of the body. The patterns chosen by the investigator reflected the patient’s clinical need for therapy, with consideration given to the severity of the involved spastic muscles of the clinical pattern, subject age/weight and muscle size, activity, and experience from previous BoNT treatments. In the bilateral group, patients were treated for pes equinus on both sides of the body (Fig. 1A). In the unilateral group, patients were treated for pes equinus and ipsilateral flexed knee or adducted thigh. In this group, patients with an AS score ? 2 in the flexed knee and/or adducted thigh had one pattern chosen for treatment based on the investigator’s judgement. Each clinical pattern was treated with half of the total incobotulinumtoxinA dose (2, 4, or 8 U/kg incobotulinumtoxinA with a maximum dose of 50, 150, and 200 U, respectively, per clinical pattern). The muscles treated for each clinical pattern are specified in Fig. 1A.

At the initial screening visit, each patient was evaluated medically for inclusion in the study, including Gross Motor Function Classification System (GMFCS) classification, AS score, and presence of pain; participants were also questioned about past and concomitant medications within the last 4 weeks, and prior BoNT-A medications. After a 14-day screening period which allowed investigators to check each subject’s eligibility for study participation, treatments were administered during two consecutive double-blind injection cycles, each followed by 12–36 weeks of observation (Fig. 1B), giving an overall study duration of 26–74 weeks. The injections were administered according to the study’s standardized treatment plans with predefined dose ranges and injection-site numbers for each muscle. Equal injection volumes were administered in all dose groups (total volume up to 8 mL; 4 mL/clinical pattern), with dose ranges and injection volumes adjusted for patients with < 25 kg BW. At least one form of technical guidance (ultrasound, electrical stimulation, or electromyography) was required for injections, and site-individualized local anesthesia and/or analgosedation protocols could be employed as needed.

Eligibility for reinjection was assessed regularly from 12–36 weeks post-injection. The treatment plan defined for the first injection cycle was continued in the second injection. Patients were eligible for re-treatment if they had an investigator- and patient-agreed clinical need for reinjection in the LL(s) and clinical patterns chosen at the injection visit of injection cycle 1, and an AS score ? 2 in the treated clinical pattern. For patients with an AS score of 1, the investigator decided whether to re-treat. The injection interval was flexible and based on clinical need. The time to reinjection for each of the three incobotulinumtoxinA dose groups was analyzed descriptively.

Participants were allowed to maintain prior usual and concomitant therapies. These included nonpharmacological therapies such as physical therapy, orthotic management other than casting and rehabilitation, and pharmacological treatments, such as muscle relaxants and antidepressants. Patients who completed the TIM study had the option of enrolling in the open-label Treatment with IncobotulinumtoxinA in Movement Open Label (TIMO) study with 4 further injection cycles.

2.3Standard protocol approvals, registrations, and patient consent

The study was conducted in accordance with the ethical principles of the Declaration of Helsinki and registered on clinicaltrials.gov (NCT01893411). The study protocol, informed consent forms, and other appropriate study-related documents were reviewed and approved by the local independent ethics committees and institutional review boards. Parents/guardians of all patients provided written informed consent, and patients provided assent if applicable.

2.4Efficacy endpoints

The primary efficacy endpoint was the change from baseline in the AS-PF on the primary body side chosen for treatment at week 4 of injection cycle 1. The 5-point AS-PF scale was used, ranging from 0 (no increase in muscle tone) to 4 (limb rigid in flexion or extension). The coprimary efficacy endpoint was the investigator’s Global Impression of Change of Plantar Flexor Spasticity Scale (GICS-PF) score on the primary body side chosen for treatment at week 4 compared with the condition before the last injection. The GICS-PF is a 7-point Likert scale from -3 (very much worse function) to +3 (very much improved function) for assessing functional change due to treatment of PF spasticity. Investigators were trained to perform and rate the AS evaluation. Based on their clinical experience-based review and judgment, investigators rated changes only of PF function compared with the condition before the last injection. Patients receiving bilateral treatment of pes equinus were assessed with the GICS-PF separately on each side. The same investigator was required to make the baseline and week 4 assessments of the first injection cycle of both coprimary endpoints and every attempt was made to have the same investigator perform the AS at all other study visits.

Secondary endpoints included AS scores for the knee flexors and thigh adductor muscles and the investigator’s Global Impression of Change Scales (GICS) to assess global changes after treatment. For the latter, overall LL spasticity change was assessed on a 7-point Likert scale from -3 (very much worse) to +3 (very much improved) and rated by the investigator, the child/adolescent (if applicable), and the parent/caregiver to gain different perspectives. The GICS has proven to be a useful assessment tool for spasticity [33].

Change in a patient’s gross motor function over time was assessed using the Gross Motor Function Measure (GMFM)-66 (scores range from 0 to 100, with higher scores reflecting better function). The GMFM-66, a shorter version of the GMFM-88, is a standardized observational instrument designed and validated to measure change in gross motor function over time in people with CP [34, 35, 36]. It has been shown to be a reliable and valid tool in children with different disabilities and is useful to assess children aged < 3 years [37].

2.5Safety

Safety endpoints assessed throughout the study included the occurrence of treatment-emergent adverse events (TEAEs), TEAEs of special interest (TEAESIs) potentially indicating distant toxin spread, and treatment-emergent serious adverse events (TESAEs). BoNT antibody testing was conducted in patients ? 21 kg BW at the screening and end-of-study visits. Blood samples were screened for BoNT antibodies using the fluorescence immunoassay for detecting antibodies (FIA-AB), and positive samples were validated using the mouse ex vivo hemidiaphragm assay (HDA) to test for neutralising antibodies (NABs).

2.6Statistical analysis

Efficacy data were analyzed descriptively in the full analysis set (a subset of patients in the safety evaluation set [SES] who had at least a baseline AS-PF score or the investigator’s GICS-PF at week 4 of injection cycle 1 available). Safety variables were analyzed descriptively in the SES, which included all patients who received at least one study treatment.

Comparison of dose groups was the primary statistical analysis. The primary and coprimary efficacy endpoints, the change from study baseline in AS-PF and investigator’s GICS-PF for the primary side at week 4, were analyzed using a mixed-model for repeated measures (MMRM; two-sided, significance level a= 0.05) or an analysis of covariance, respectively. Least squares mean values were compared using a two-step hierarchical test procedure to detect differences between incobotulinumtoxinA high and low dose in step 1, and between mid and low dose in step 2. If one of two hierarchical tests did not yield a statistically significant result, the consecutive test(s) were still performed but were considered descriptive only.

Secondary and other analyses included the change from baseline in AS-PF and investigator’s GICS-PF for the primary side at week 4 of injection cycle 2 and at additional times during injection cycle 1 and injection cycle 2 and changes from baseline in secondary endpoints at various times during both injection cycles. Statistical analyses were performed using the statistical analysis system (SAS®) software package, version 9.2 or higher (SAS Institute, Cary, NC, USA).

Safety data were analyzed descriptively.

2.7Sample size calculations and study power

It was estimated that the sample size of N= 300 participants would provide 94% power (product of the single power calculations for both the primary and the coprimary efficacy variables for the high- vs. low-dose treatment comparison) to identify a statistically significant difference between the high- and low-dose groups. A power of 83.3% was provided for the treatment comparison of the mid versus the low dose in both primary efficacy variables. Using a randomization ratio of 2:1:1, at least 150 subjects were planned to be randomized to the high-dose treatment group and at least 75 subjects each were planned to be randomized to the mid- and low-dose treatment groups.

Figure 2.

Patient disposition. *Multiple entries possible. **Subjects who completed IC1 and continued to IC2. BW = body weight; IC = injection cycle; kg = kilogram; U = unit.

2.8Data availability statement

Key elements of the study protocol, design, and statistical analysis plan were deposited in the U.S. National Library of Medicine database (www.clinicaltrials.gov, NCT01893411) and EU Clinical Trials Register (https: //eudract.ema.europa.eu/, 2012-005054-30). All relevant information is contained within this manuscript and the supplementary material.

3.1Efficacy

3.1.2Further results

AS

Consistent improvements for all patterns with each injection cycle were seen at weeks 4 and 8. AS-PF scores of the primary body side further improved from baseline to week 4 of the second injection cycle in all three dose groups (Fig. 3). The treatment effects were greatest 4 and 8 weeks following each injection, after which changes decreased by week 12. Following unilateral LL treatment, all doses of incobotulinumtoxinA resulted in improvements from baseline in AS scores of the knee flexors (Fig. 5A) and the thigh adductors (Fig. 5B) 4 weeks following each injection. Wherever estimable, all doses of incobotulinumtoxinA produced comparable improvements.

Figure 5.

The effect of incobotulinumtoxinA on mean change from baseline on week 4 on AS as measured on the (A) knee flexors and (B) thigh adductor muscles, FAS, OC. **p< 0.05 versus study baseline. AS = Ashworth Scale; BW = body weight; FAS = full analysis set; IC = injection cycle; kg = kilogram; OC = observed cases; SE = standard error; U = unit.

Investigator’s, child/adolescent’s, and parent/caregiver’s GICS scores

Whether from the perspective of the investigator, the parent/caregiver, or the patient, GICS scores confirmed a consistent and global improvement in LL spasticity at 4 weeks post-injection for all three incobotulinumtoxinA dose groups and across both injection cycles (Fig. 6A–C). Across both injection cycles, investigators indicated improvement rates of > 90% when taking the sum of “minimally improved,” “much improved,” and “very much improved” frequencies. Parents/caregivers reported improvements in more than 80% of patients at these times. GICS scores were available from 150 of 311 and 135 of 287 children/adolescents (48% and 47.0%) at injection cycle 1 and injection cycle 2, respectively; of those who responded, > 80% reported improvement after each injection cycle.

Motor functioning

GMFM-66 scores indicated that motor function improved in all three treatment groups from the end of injection cycle 1 to the end-of-study visit 8 weeks later. Mean (SD) GMFM-66 scores increased from baseline by 1.8 (2.8), 1.2 (3.5), and 1.4 (3.1) at the end of injection cycle 1 and by 3.1 (3.4), 3.3 (4.5), and 2.8 (4.1) at the end-of-study visit in patients who received incobotulinumtoxinA 4, 8, and 16 U/kg (maximum 100, 300, and 400 U), respectively.

3.2Safety

IncobotulinumtoxinA treatment was generally well-tolerated over both injection cycles, with TEAEs reported in 42.8% of patients overall (Table 2). The incidence of TEAEs was slightly higher in patients who received the highest dose of incobotulinumtoxinA, followed by the lowest dose, and then the mid dose. Most TEAEs were mild or moderate in intensity. TEAEs assessed by investigators as treatment-related also occurred at a low frequency (4.8% of patients overall). These events were localized muscular weakness (n= 5 patients), injection-site pain (n= 4 patients), injection-site erythema, pain in an extremity, pyrexia (n= 2 patients each), fall, hematoma, influenza-like illness, injection-site inflammation, injection-site warmth, and rash (n= 1 patient each).

Figure 6.

The effect of incobotulinumtoxinA on (A) investigator, (B) parent/caregiver, and (C) child/adolescent GICS scores† at week 4, FAS, OC. †GICS scores were available from 150 of 311 and 135 of 287 children/adolescents (48% and 47.0%) at IC1 and IC2. The proportion of children/adolescents responding was attributed to the respondents’ young age or their cognitive abilities. ***P< 0.0001. BW = body weight; FAS = full analysis set; GICS = Global Impression of Change Scale; IC = injection cycle; kg = kilogram; OC = observed cases; SE = standard error; U = unit.

TEAESIs were reported in 2.3% of patients overall. The most frequent TEAESI, localized muscular weakness, was reported in five patients and, in all cases, affected the treated LL and was considered by the investigator to be related to treatment. Given the low incidence of TEAESIs, no meaningful conclusions could be drawn concerning any relationship between CP severity and TEAESIs.

TESAEs occurred at low frequencies in all dose groups for each injection cycle (high dose: 2.6% and 2.1%; mid dose: 0.0% and 1.4%; low dose: 3.8% and 4.1% in the first and second cycles, respectively). No treatment-related TESAE or fatal TEAE was reported. One patient in the high-dose group discontinued the study because of a non-treatment-related nonserious TEAE (asthma).

At the screening visit, 19 of 127 (15.0%) subjects eligible for testing were tested positive on the FIA-AB. Three patients subsequently tested positive for NABs according to HDA test results; all three patients had been pretreated with other BoNT-As. Of these, one had a negative HDA test at the end-of-study visit, and the other two patients did not undergo further HDA testing. Based on a protocol deviation, ten patients at screening and two patients at the end-of-study visit were FIA-AB positive but were not HDA tested (HDA missing). All patients with positive HDA or HDA missing based on protocol deviation responded to treatment, based on the investigator’s GICS-PF assessment.

4.1Strengths and limitations

The strengths of the TIM study include its patient-centric approach to treatment. Individualized treatment options within standardized, multipattern treatment protocols for LL spasticity were available to address the real-world needs of a large, heterogeneous population of children and adolescents with CP. Participants represented a broad age range (2–17 years) and all levels of disease severity (GMFCS-E&R levels I–V). The efficacy and safety of incobotulinumtoxinA were assessed across a wide dose range. The study design allowed for the treatment of several muscle groups at once, giving physicians the freedom to adjust dose, injection site, number of injections per site, and time to reinjection as required, based on individual patient need. Additional strengths were the generation of a comprehensive dataset, which included efficacy and safety data for other LL muscle groups in addition to the PFs and the uni/bilateral combination treatment. The outcomes provided data on the effect of incobotulinumtoxinA on muscle resistance, global improvement, and gross motor function, from the point of view of the patient, parent/caregiver, and physician.

The lack of a placebo control in the TIM study may be viewed as a limitation, but use of placebo in this vulnerable young patient population raises concerns in terms of ethical aspects and feasibility. Furthermore, the effect sizes, as discussed earlier, were within established ranges, and the current patient-centric approach allowed for a thorough investigation of long-term efficacy and safety across a broad dose range of incobotulinumtoxinA in children and adolescents with spastic CP.