Journal of Neurology

, Volume 259, Issue 2, pp 284–291

Assessment of neurological efficacy of idebenone in pediatric patients with Friedreich's ataxia: data from a 6-month controlled study followed by a 12-month open-label extension study

Authors

  • Thomas Meier
    • Santhera Pharmaceuticals
  • Susan L. Perlman
    • Department of NeurologyDavid Geffen School of Medicine at the University of California at Los Angeles
  • Christian Rummey
    • Santhera Pharmaceuticals
  • Nicholas J. Coppard
    • Santhera Pharmaceuticals
    • Division of NeurologyThe Children’s Hospital of Philadelphia
Original Communication

DOI: 10.1007/s00415-011-6174-y

Cite this article as:
Meier, T., Perlman, S.L., Rummey, C. et al. J Neurol (2012) 259: 284. doi:10.1007/s00415-011-6174-y

Abstract

The aim of this study was to investigate the efficacy of idebenone on neurological function as assessed by ICARS and FARS neurological rating scales in pediatric Friedreich's ataxia (FRDA) patients. Sixty-eight pediatric patients were enrolled in an open-label extension study (IONIA-E) where patients received idebenone (Catena®, 150 mg film-coated tablets) at a weight-adjusted dose of 1,350/2,250 mg/day for 12 months after patients had completed a double-blind, randomized, placebo-controlled study (IONIA) receiving either idebenone at a weight-adjusted dose of 450/900 or 1,350/2,250 mg/day or placebo for 6 months. Changes in ICARS and FARS total scores and subscores were recorded for the 12-month IONIA-E study and for the 18-month combined IONIA and IONIA-E study period. Data analyzed by a mixed-model repeated-measures ANCOVA relative to baseline resulted in least square means for the change in ICARS for the IONIA-E study of +0.98 points (SEM 0.73; p = 0.180), indicating a trend for worsening. However, combined with the IONIA study the change was −1.03 ± 0.68 points (p = 0.132), indicating a trend for improvement in neurological function over the 18-month period. Importantly, patients who received idebenone 1,350/2,250 mg/day over this period significantly improved in neurological function (change in ICARS: −3.02 ± 1.22, p = 0.014). The improvement in neurological function over time was best seen when the posture and stance subscore was excluded from the analysis. Comparable data were obtained with the FARS. The findings of the open-label IONIA-E study combined with the double-blind IONIA study indicate that idebenone at a dose of 1,350/2,250 mg/day may offer a therapeutic benefit to pediatric FRDA patients by stabilizing the overall neurological function and improving fine motor skills and speech.

Keywords

AtaxiaInternational Cooperative Ataxia Rating ScaleIdebenoneMovement disorder

Introduction

Friedreich's ataxia (FRDA) is a severely debilitating neurodegenerative disease that has a major influence on the lives of affected individuals. The condition has an autosomal recessive mode of inheritance and neurological symptoms are characterized by progressive gait and limb ataxia, dysarthria, lower-limb areflexia, decreased vibration sense, and muscular weakness [13]. Patients with FRDA usually present around the time of puberty with in-coordination of movements and dysarthria. Progressive limb and gait ataxia occurs, with most patients being confined to a wheelchair by their late 20s [4]. Life expectancy is reduced to an average of between 30 and 40 years [5]; however, patients sometimes survive well above their sixth decade.

Pathogenic trinucleotide (GAA) repeat expansions in the first intron of the FXN gene cause reduced frataxin expression, which in turn results in deficient assembly of iron–sulfur clusters, elevated oxidative stress, and impaired cellular energy production [6]. Idebenone, a short-chain benzoquinone, acts as an antioxidant and ATP production modulator and has been investigated extensively as a potential symptomatic treatment for FRDA [7, 8]. Efficacy of idebenone on neurological symptoms in pediatric FRDA patients, as assessed by ataxia rating scales, has been demonstrated in one randomized, double-blind, placebo-controlled study [9] and in several open-label studies [1012]. However, a recently completed randomized, controlled 6-month study in pediatric patients did not confirm the efficacy of idebenone on neurological symptoms, (the IONIA study; [13]), mainly due to the modest improvement in neurological symptoms detected in the idebenone treated group combined with the unexpected improvement seen in the placebo group. To investigate further the potential efficacy of idebenone on the development of neurological symptoms in pediatric FRDA patients, the course of neurological function was followed in 68 patients who participated in the 6-month IONIA study followed by a 12-month open-label extension (IONIA-E) study during which all patients received high-dose idebenone.

Methods

Objective

This report analyzes the efficacy of idebenone on neurological function in pediatric patients enrolled in a double-blind, randomized, placebo-controlled, parallel group study (IONIA; clinicaltrials.gov ID: NCT00537680) of 6-month duration followed by an open-label extension study (IONIA-E; clinicaltrials.gov ID: NCT00697073) of 12-month duration.

Patients

Seventy genetically confirmed pediatric patients enrolled and completed the 6-month IONIA study [13]. Patients completing the IONIA study with a body weight ≥25 kg and able to comply with study procedures were eligible to enroll in the IONIA-E study. Exclusion criteria included pregnancy/breast-feeding, clinically significant abnormalities in clinical hematology or biochemistry, and abuse of drugs or alcohol. Sixty-eight patients elected to enroll in the open-label IONIA-E study (two patients decided not to enroll, due to non-medical reasons). An equal number of patients were enrolled at The Children’s Hospital of Philadelphia and the University of California at Los Angeles. Data from these 68 patients are analyzed for the 12-month (52-week) IONIA-E study and separately for the 18-month (76-week) period of the combined IONIA and IONIA-E studies.

Treatment allocation

During the IONIA study, patients were randomized 1:1:1 to one of three treatment arms: group A received idebenone (Catena®, 150 mg tablets) at 450 mg/day (if ≤45 kg body weight at baseline) or 900 mg/day (if >45 kg body weight at baseline); group B received idebenone 1,350 mg/day (if ≤45 kg) or 2,250 mg/day (if >45 kg); and group C received placebo. During the randomized IONIA study, 22 patients each received idebenone dose A and B, respectively; 24 patients received placebo. All patients received idebenone dose B during the IONIA-E study. Study medication was administered in three divided doses together with meals.

Study conduct

Visit 1/day 1 of the IONIA-E study took place no more than 27 days after the last visit of the 24-week IONIA study. During the IONIA-E study, subjects were seen by the investigator at day 1, weeks 4, 12, 25, and 52 and approximately 4 weeks upon discontinuation of study medication intake. Patients maintained a patient diary, with daily documentation of study medication intake and recording of adverse events and concomitant medications. All patients signed informed consent forms for both studies, which had ethics approvals from both study sites and were conducted according to GCP and ICH guidelines.

Outcome measures, data evaluation, and statistics

Neurological function was assessed using the International Cooperative Ataxia Rating Scale, ICARS [14] and Friedreich Ataxia Rating Scale, FARS [15]. Scale items were grouped into functional domains, deviating slightly from the original descriptions, to allow across-scale comparisons and interpretations (Supplementary Table for details). Statistical analyses were based on the intent-to-treat population using observed cases. For the open-label IONIA-E study, the assessment of neurological efficacy by ICARS and FARS scales was planned without prioritization of endpoints. According to the study protocol and the prospectively defined statistical analysis plan, no correction for multiplicity was applied. The analysis of the combined IONIA and IONIA-E studies was prespecified as exploratory analysis. For the IONIA-E study, the change in absolute ICARS and FARS values from baseline to the last visit was analyzed using a repeated-measures ANCOVA model with visit, treatment (randomized treatment group in the IONIA study), and visit by treatment interaction as factors and baseline as covariate. In addition, data from the IONIA study was integrated with IONIA-E study data and summarized by time point and the change from the IONIA baseline to the last visit of the IONIA-E study (18-month period) was analyzed in the same way.

Results

The mean age of the 68 pediatric subjects included was 13.5 years at the time of enrollment in the IONIA study (Table 1); by the time patients were enrolled into the open-label extension study, three subjects were older than 18 years (maximum age 18.6 years). There were no imbalances across the study centers with respect to demographic parameters (data not shown). Nine patients did not complete the IONIA-E study according to the study protocol (three patients for non-compliance with study medication, two patients withdrew consent, one for pregnancy, and three for other reasons). The mean compliance with study drug was 94.3% (SD 9.2%) as assessed by pill count.
Table 1

Patient demographics

Category

At baseline

IONIA

IONIA-E

Age (years)

 Mean (SD)

13.5 (2.73)

14.0 (2.72)

 Median (range)

13.5 (8–18)

14.0 (8.5–18.6)

Sex, n (%)

 Male

32 (47.1)

Nc

 Female

36 (52.9)

 

GAA repeats, n (%)

 ≤800

50 (73.5)

Nc

 >800

18 (26.5)

 

Baseline ICARS

 Mean (SD)

35.1 (7.63)

33.1 (9.70)

 Median (range)

34 (16–52)

32.0 (8–55)

Baseline FARS

 Mean (SD)

56.9 (10.1)

56.3 (12.3)

 Median (range)

57.5 (33.8–81)

56.0 (29–81)

Disease duration (months)

 Mean (SD)

66.12 (38.40)

Nc

 Median (range)

67.98 (0.8–196.9)

 

Nc No change to IONIA baseline

The objective of this study was to follow the course of the neurological symptoms under treatment with idebenone. The ICARS and FARS total scores at baseline of the IONIA study and at the baseline of the IONIA-E study (Table 1) were comparable across the treatment groups (not shown).

For the 12-month IONIA-E study, the mean ICARS increased by 1.1 points (SD 5.64), indicating a slight worsening (Fig. 1), which contrasts with the mean decrease of 2.0 points (SD 5.27) for the 6-month IONIA study (Fig. 2a). Considering the entire 18-month study period, the mean ICARS decreased by 1.2 points (SD 6.05) from the baseline of the IONIA study to week 52 of the extension study (Fig. 2a). Applying the mixed-model repeated-measures ANCOVA resulted in an estimated mean increase of 0.98 ICARS points (SEM 0.73, p = 0.180) from baseline to week 52 of the IONIA-E study, which did not reach statistical significance (Table 2). In contrast, there was a decrease in estimated mean by 1.03 ICARS points (SEM 0.68, p = 0.132) for the 18-month period of the combined IONIA and IONIA-E studies.
https://static-content.springer.com/image/art%3A10.1007%2Fs00415-011-6174-y/MediaObjects/415_2011_6174_Fig1_HTML.gif
Fig. 1

a Change in ICARS total score for the 12-month open-label IONIA-E study (baseline to week 25 and week 52). b Change in ICARS posture and stance subscore. c Change in ICARS subscores excluding the posture and stance subscore. Data is raw means ± SEM. W weeks

https://static-content.springer.com/image/art%3A10.1007%2Fs00415-011-6174-y/MediaObjects/415_2011_6174_Fig2_HTML.gif
Fig. 2

a Change in ICARS total score for the 18-month combined study period (IONIA study baseline to week 52 of the IONIA-E study). b Change in ICARS posture and stance subscore. c Change in ICARS subscores excluding the posture and stance subscore. Data is raw means ± SEM. W weeks

Table 2

Change in ICARS total score and subscores by study period

 

IONIA-E baseline to IONIA-E week 52 (12-month study period)

IONIA baseline to IONIA-E week 52 (18-month study period)

Changea

I/W

p value

Changea

I/W

p value

ICARS total score

+0.98 ± 0.73

W

0.180

−1.03 ± 0.68

I

0.132

Eye subscore

−0.08 ± 0.09

I

0.356

−0.37 ± 0.08

I

<0.0001

Speech subscore

−0.33 ± 0.11

I

0.003

−0.80 ± 0.10

I

<0.0001

Upper limb ataxia subscore

−0.32 ± 0.32

I

0.316

−1.21 ± 0.28

I

<0.0001

Lower limb ataxia subscore

−0.08 ± 0.26

I

0.760

−1.27 ± 0.22

I

<0.0001

Posture and stance subscore

+1.75 ± 0.35

W

<0.0001

+2.70 ± 0.36

W

<0.0001

Spiral subscore

+0.03 ± 0.09

W

0.758

−0.07 ± 0.07

I

0.329

I Improvement, W worsening, IONIA-E IONIA extension study

aScore estimated mean changes ±SEM from ANCOVA model

The analysis for each of the ICARS subscores showed that the posture and stance subscore (Supplementary Table) changed in a different pattern compared to the remaining subscores. This was the only ICARS subscore for which the entire study population showed a constant, approximately linear and statistically significant increase during the IONIA-E study (p < 0.0001) as well as during the 18 months of the combined study period (p < 0.0001), representing a functional decline (Table 2; Fig. 2b). However, the overall increase in the posture and stance subscore was statistically significant only for the placebo (p < 0.0001) and idebenone 450/900 mg/day groups (p < 0.0001) and was numerically smaller and not significant for the idebenone 1,350/2,250 mg/day group (p = 0.065) for the combined study period.

In contrast, excluding the posture and stance subscore from the analysis resulted in a constant decrease in ICARS for the sum of the eye, speech, upper and lower limb ataxia, and spiral subscores for the duration of the IONIA-E study (p = 0.155) as well as for the combined duration of the IONIA and IONIA-E study (p < 0.0001), representing an improvement in function for these combined subscores (Fig. 1c). This was statistically significant for the speech subscore of the IONIA-E study and for the eye, speech, upper and lower limb ataxia subscores for the combined study periods of the IONIA and IONIA-E studies (Table 2).

Separating the total ICARS score by treatment group showed that patients initially randomized to receive 1,350/2,250 mg/day idebenone during the IONIA study benefited most, as they ended the combined 18-month observation period with the lowest ICARS score (Fig. 3). Applying the ANCOVA model of repeated measures, patients initially randomized to the idebenone 1,350/2,250 mg/day group showed a statistically significant decrease from baseline in estimated means for the total ICARS by 3.02 points (SEM 1.22, p = 0.014). The idebenone 450/900 mg/day group had an estimated mean increase of 0.11 ICARS points (SEM 1.19, p = 0.93) and the placebo group decreased by 0.18 ICARS points (SEM 1.14, p = 0.88), indicating that at the end of the IONIA-E study the total ICARS scores in these two groups were not statistically different from the ICARS at the beginning of the IONIA study. Accordingly, the estimated differences in ICARS total scores between treatment groups for the 18 months of the combined study period as determined by the ANCOVA model revealed trends towards a difference between patients in the idebenone 1,350/2,250 mg/day group and the placebo group (difference in estimated means for the total ICARS: −2.83 points, SEM 1.67, p = 0.091) and patients in the idebenone 450/900 mg/day group (difference in estimated means for the total ICARS: −3.12, SEM 1.71, p = 0.070). A responder analysis indicated that 63.9% of all patients showed improvement or no worsening on the ICARS, while 31.1% of patients improved by at least 5 ICARS points over the 18-month combined study period (Table 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs00415-011-6174-y/MediaObjects/415_2011_6174_Fig3_HTML.gif
Fig. 3

Change in ICARS total score for the 18-month combined study period by treatment group as assigned at the baseline of the IONIA study. Data is raw means ± SEM. W weeks

Table 3

Responder analysis for all patients and separated according to treatment group as assigned at baseline of the IONIA study

Group

Number of subjects, n (%)

Decrease by ≥0 ICARS pointsa

Decrease by ≥2.5 ICARS pointsb

Decrease by ≥5 ICARS pointsc

All patients

39 of 61 (63.9%)

27 of 61 (44.3%)

19 of 61 (31.1%)

Placebod

11 of 20 (55%)

8 of 20 (40%)

5 of 20 (25%)

Idebenone dose A 450/900 mg/dayd

15 of 19 (78.9%)

11 of 19 (57.9%)

9 of 19 (47.4%)

Idebenone dose B 1,350/2,250 mg/dayd

13 of 22 (59.1%)

8 of 22 (36.4%)

5 of 22 (22.7%)

Number and percentage of subjects with no change or any improvement in FRDA status as evidenced by no change or decrease in the total ICARS for the total study period of 18 months. Only subjects who have completed the indicated study period are included

aNumber and percentage of subjects with no change or any decrease (≥0 points)

bSubjects with ≥2.5 points decrease in ICARS

cSubjects with ≥5 points decrease in ICARS

dTreatment as assigned at baseline of the IONIA study

The changes in FARS total score and subscores followed a similar pattern to that of the ICARS. Applying the ANCOVA model of repeated measures to the IONIA-E study resulted in an estimated mean increase of 2.23 total FARS points (SEM 0.78; p = 0.005; Table 4) for all patients. For the entire 18-month combined study period, there was an estimated mean increase of 1.58 total FARS points (SEM 0.74; p = 0.034) (Table 4). Comparable to the differential change in ICARS subscores, there were also differences in the changes of individual FARS subscores with a similar pattern for the IONIA-E study and for the combined IONIA and IONIA-E studies (Table 4). The most improvement was seen for the speech subscore (both for the period of the IONIA-E study as well as for the combined IONIA and IONIA-E study periods), which is aligned with the changes in the speech subscore of the ICARS. In contrast, the stability and gait subscore and the PNS subscore increased, representing a deterioration in function. In agreement with the ICARS data, the treatment group receiving idebenone at a dose of 1,350/2,250 mg/day for the entire 18-month study period performed best on the FARS (Fig. 4). Specifically, applying the ANCOVA model for each of the treatment groups over the 18-month combined study period, there was a trend towards a difference between patients in the idebenone 1,350/2,250 mg/day group and the placebo group of −3.17 FARS points (SEM 1.81, p = 0.080). In contrast, no difference was seen between patients in the idebenone 1,350/2,250 mg/day group and patients in the idebenone 450/900 mg/day group (difference in total FARS: −0.99, SEM 1.85, p = 0.59).
Table 4

Change in FARS total score and subscores by study period

 

IONIA-E baseline to IONIA-E week 52 (12-month study period)

IONIA baseline to IONIA-E week 52 (18-month study period)

Changea

I/W

p value

Changea

I/W

p value

FARS total score

+2.23 ± 0.78

W

0.005

+1.58 ± 0.74

W

0.034

Bulbar subscore

−0.01 ± 0.04

I

0.740

+0.02 ± 0.04

W

0.614

Speech subscore

−0.15 ± 0.06

I

0.020

−0.26 ± 0.06

I

<0.0001

PNS subscore

+0.81 ± 0.28

W

0.005

+1.18 ± 0.25

W

<0.0001

Upper limb coordination subscore

+0.12 ± 0.39

W

0.750

−0.28 ± 0.38

I

0.472

Lower limb coordination subscore

−0.04 ± 0.22

I

0.856

−0.63 ± 0.20

I

0.002

Stability and gait subscore

+1.49 ± 0.36

W

<0.0001

+1.55 ± 0.33

W

<0.0001

I Improvement, W worsening, IONIA-E IONIA extension study

aScore estimated mean changes ±SEM from ANCOVA model

https://static-content.springer.com/image/art%3A10.1007%2Fs00415-011-6174-y/MediaObjects/415_2011_6174_Fig4_HTML.gif
Fig. 4

Change in FARS total score for the 18-month combined study period by treatment group as assigned at the baseline of the IONIA study. Data is raw means ± SEM. W weeks

Discussion

The current study summarizes data from a 12-month open-label extension study (IONIA-E) in pediatric subjects with FRDA, presented as stand-alone data as well as in conjunction with data from a preceding 6-month randomized, placebo-controlled clinical trial (IONIA). Combining data from all patients for the IONIA-E study resulted in a statistically not significant minimal worsening in total ICARS and a statistically significant worsening in total FARS. Extending the observational period to the 18 months of the combined IONIA and IONIA-E study resulted in a non-significant improvement in total ICARS and a significant worsening in total FARS.

A striking finding of this study was the observation that the posture and gait subscore of the ICARS constantly increased (worsened) in an almost linear pattern over the period of the IONIA and IONIA-E studies, representing deterioration in function. In agreement with the finding for the posture and stance subscore of the ICARS, there was also a constant increase in the stability and gait subscore of the FARS (the functional equivalent of the ICARS posture and stance subscore). In addition, the PNS subscore of the FARS also increased. Although patients markedly worsened in the posture and stance subscore of the ICARS, they remained stable or improved in all ICARS subscores excluding the posture and stance subscore both during the IONIA-E study period as well as for the combined IONIA and IONIA-E study periods. For the IONIA-E study, the improvement was particularly evident in the speech subscore, while for the combined IONIA and IONIA-E period, speech as well as the upper and lower limb ataxia subscores of the ICARS improved. A comparable pattern was seen for the FARS with consistent improvements seen in the speech subscore. This finding is in agreement with previous reports, which indicated that idebenone might be most efficacious in altering the fine motor skills of FRDA patients, especially speech [10, 11]. The pronounced improvement of speech function is of particular interest, as FRDA patients clearly develop some degree of dysarthria early in the course of the disease and typically at a young age. In the future, more sophisticated language-assessment strategies [16, 17] might help further to analyze the potential of idebenone treatment for this aspect of the disease, which has a clear impact on the patients’ quality of life. This might allow the use of speech- and hearing-related outcome measures as medically relevant endpoints for future clinical trials.

Separating the change in total ICARS and FARS scores for the IONIA-E study by treatment allocation during the preceding 6-month IONIA study demonstrated that the neurological function assessed by the ICARS and FARS scales improved for patients who received a high dose 1,350/2,250 mg/day idebenone throughout the combined studies. Specifically, patients continuously receiving high-dose idebenone decreased by ~1 point on the ICARS during the IONIA-E study, which contrasts with the expected natural rate of disease progression represented by an increase of ~4–5 ICARS points per year described in observational studies for adult patients [18, 19]. Patients who had continuously received high-dose (1,350/2,250 mg/day) idebenone for the entire period of 18 months improved by more than 3 points on the ICARS and also performed best on the FARS. This finding could be interpreted to represent a neuro-protective effect; however, this result would require confirmation by an independent study, possibly applying a delayed-start design.

The observed difference on the ICARS cannot be explained by an age-dependent improvement on ICARS ratings described for young children [20] as there was a clear treatment-dependent difference in the change of the ICARS favoring the group of patients who continuously received high-dose idebenone. In addition, patients who had been on placebo or 450/900 mg/day idebenone for the 6-month IONIA study and who only received idebenone 1,350/2,250 mg/day for the IONIA-E study experienced a slight worsening for this 12-month open-label treatment period but did not deteriorate over the 18-month period of the combined IONIA and IONIA-E studies. Overall, this relative stabilization in neurological function also can be considered a treatment effect for this progressive neurodegenerative disease, as children with FRDA generally progress faster than adults, making the improvement in all groups relative to the natural history more substantial [18, 21].

The interpretation of the results from the IONIA-E study is complicated by several factors, including the short-term improvement on ICARS and FARS in the placebo group during the preceding IONIA study [13]. Additional confounding factors might be the use of a restricted study cohort possibly presenting a selection bias, and potential practice effects due to the frequent ICARS and FARS assessments. Findings presented here should also be interpreted with caution, as the assessments are subject also to day-to-day variability of the patients’ functional capabilities and in light of the limited observational period and sample size.

Despite these limitations, the findings of the open-label IONIA-E study combined with the data from the double-blind, randomized IONIA study indicate that idebenone at a dose of 1,350/2,250 mg/day may offer a therapeutic benefit to FRDA patients by stabilizing the overall neurological function with the potential of actually improving upper and lower limb coordination and speech. Combined with the well-established safety and tolerability [9, 13, 22], idebenone currently remains to be an ongoing treatment option for pediatric FRDA patients.

Acknowledgments

We would like to acknowledge assistance of the clinical coordinators of the study: Lisa Friedman, Erin Paulsen, Lynn Kessler, Sharaone Trifskin, and Bonnie Johnson. Statistical analyses were conducted by an independent statistician, Britt-Marie Lindström, 4Pharma, Sweden. The present study was sponsored by Santhera Pharmaceuticals. Dr. Lynch has received grant funding for other projects from the National Institutes of Health, the Muscular Dystrophy Association, the Friedreich Ataxia Research Alliance and Penwest Pharmaceuticals. Dr. Perlman has also received grant funding for other projects from the National Institutes of Health, the HighQ/CHDI Foundation, the National Ataxia Foundation, the Muscular Dystrophy Association, the Friedreich Ataxia Research Alliance, and Edison Pharmaceuticals. Drs. Meier, Coppard and Rummey are employees of Santhera Pharmaceuticals.

Conflict of interest

None.

Supplementary material

415_2011_6174_MOESM1_ESM.doc (36 kb)
Supplementary material 1 (DOC 36 kb)

Copyright information

© Springer-Verlag 2011