-1-
Progression characteristics in Friedreich’s Ataxia: 4-year analysis of the
European Friedreich’s Ataxia Consortium for Translational Studies (EFACTS)
Kathrin Reetz MD
1,2
*, Imis Dogan PhD
1,2
*, Ralf-Dieter Hilgers PhD
3
, Paola Giunti MD
4
, Michael H
Parkinson MD
4
, Caterina Mariotti MD
5
, Lorenzo Nanetti MD
5
, Alexandra Durr MD
6
, Claire Ewenczyk
6
,
Sylvia Boesch MD
7
, Wolfgang Nachbauer MD PhD
7
, Thomas Klopstock MD
8,9,10
, Claudia Stendel
MD
8,9
, Francisco Javier Rodríguez de Rivera Garrido MD
11
, Christian Rummey
12
, Ludger Schöls
MD
13,14
, Stefanie Hayer PhD
13
, Thomas Klockgether MD
15,16
, Ilaria Giordano MD
15
, Claire Didszun
PhD
1
, Myriam Rai PhD
17
, Massimo Pandolfo MD
17
, Jörg B. Schulz MD
1,2
and the EFACTS study group
#
*Shared authorship;
#
Listed at the end of the paper
1
Department of Neurology, RWTH Aachen University, Pauwelsstraße 30, Aachen, Germany
2
JARA-BRAIN Institute Molecular Neuroscience and Neuroimaging, Forschungszentrum Jülich GmbH
and RWTH Aachen University, 52074 Aachen, Germany
3
Department of Medical Statistics, RWTH Aachen University, Pauwelsstraße 19, Aachen, Germany
4
Ataxia Centre, Department of Clinical and Movement Neurosciences, UCL-Queen Square Institute
of Neurology, London, United Kingdom
5
Unit of Medical Genetics and Neurogenetics, Fondazione IRCCS Istituto Neurologico Carlo Besta,
Milan, Italy
6
Sorbonne Universit, Paris Brain Institute (ICM Institut du Cerveau), AP-HP, INSERM, CNRS,
University Hospital Piti-Salptrire, Paris, France
7
Department of Neurology, Medical University Innsbruck, Austria
8
Department of Neurology, Friedrich Baur Institute, University Hospital, LMU, Munich, Germany
9
German Center for Neurodegenerative Diseases (DZNE), Munich, Germany
10
Munich Cluster for Systems Neurology (SyNergy), Munich, Germany
11
Reference Unit of Hereditary Ataxias and Paraplegias, Department of Neurology, IdiPAZ, Hospital
Universitario La Paz, Madrid, Spain
12
Clinical Data Science GmbH, Basel, Switzerland
13
Department of Neurology and Hertie-Institute for Clinical Brain Research, University of Tübingen,
Tübingen, Germany
14
German Center for Neurodegenerative Diseases (DZNE), Tübingen, Germany
15
Department of Neurology, University Hospital of Bonn, Bonn, Germany
16
German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany
17
Laboratory of Experimental Neurology, Université Libre de Bruxelles, Brussels, Belgium
Corresponding author:
Prof Dr. Kathrin Reetz
Department of Neurology
RWTH Aachen University
Pauwelsstrasse 30
52074 Aachen, Germany
Phone: +49-241-80 89600
Fax: +49-241-80 3336516
E-mail: kreetz@ukaachen.de
Summary: 298/300 Article 3.500/3.500 References: 30/30
-2-
Summary
Background The European Friedreich’s Ataxia Consortium for Translational Studies
(EFACTS) investigates the natural history of Friedreich’s ataxia (FRDA). We aimed to assess
progression characteristics of clinical rating scales, functional patient-reported measures and
performance-based coordination tests based on longitudinal four-year data.
Methods EFACTS is prospective, longitudinal, observational cohort study based on an
ongoing and open-end registry. Patients with genetically confirmed FRDA are enrolled at 11
clinical centres in seven European countries and seen on a yearly basis. Data of up to five
visits from baseline to four-year follow-up was included in the current analysis. Our primary
endpoints were the clinical Scale for the Assessment and Rating of Ataxia (SARA) and the
functional Activities of Daily Living (ADL) scale. Annual disease progression was analysed with
linear mixed effect models and additional regression statistics for detailed subgroup
characterization. This study is registered at https://clinicaltrials.gov (NCT02069509).
Findings In total, 602 FRDA patients, assessed between 15-Sep-2010 and 05-Mar-2018,
were included. Of these, 552 patients (92%) contributed data with at least one follow-up visit.
Annual progression rate for SARA was 0·82 points (SE 0·05) in the overall cohort, and higher
in ambulatory (1·12 [0·07]) compared to non-ambulatory (0·50 [0·07]) patients. Subitem-
analyses revealed high rates of progression for lower limb components in ambulatory patients.
ADL worsened with 0·93 (0·05) points per year, with subitem falls showing strongest effects
over time. For a two-year parallel-group clinical trial, about 118 (59 per group) ambulatory
patients are required to detect a 50% reduction in SARA progression at 80% power. Less
patients are needed using ADL as a functional outcome in a similar design and including only
early-onset ambulatory individuals.
Interpretation Our findings on disease-stage dependent clinical and functional progression
have important implications for clinicians and researchers, and enable tailored sample size
calculation to guide upcoming trial-designs in FRDA.
Funding FP7 Grant from the European Commission (HEALTH-F2-2010- 242193), Voyager
Therapeutics and Euro-Ataxia.
-3-
Introduction
Friedreich’s ataxia (FRDA) is an autosomal-recessive multisystem disorder characterized by
spinocerebellar ataxia, dysarthria, pyramidal weakness, deep sensory loss, hypertrophic
cardiomyopathy, skeletal abnormalities, and diabetes mellitus
1,2
. This chronic progressive
neurodegenerative disease has a typical onset around puberty, and is a consequence of a
deficiency of the protein frataxin, a mitochondrial protein involved in iron sulfur cluster
synthesis. Most patients are homozygous for the hyperexpansion of a guanosine-adenosine-
adenosine (GAA) repeat in the first intron of the FXN gene
3
. By early adulthood – often after
about 15 years of disease manifestation – patients with FRDA are bound to wheelchair
4
. A
major cause of death is cardiomyopathy
5
.
Currently, there is no cure and no approved treatment for FRDA. However, several potential
disease-modifying treatments in FRDA are emerging. These therapeutic strategies aim to
restore FXN levels either by upregulating the endogenous gene, or by protein or gene
replacement therapies. Translating these approaches into clinical trials and practice requires
sound clinical trial designs. This is particularly challenged by (i) the low incidence of the
disease, (ii) the search for the most meaningful and sensitive measure(s) in view of the slow
progression in FRDA, and (iii) the complexity of this multi-organ disease.
We recently reported baseline
6
and two-year follow-up
7
data about the neurological and
functional status of FRDA patients from the prospective registry of the European Friedreich’s
Ataxia Consortium for Translational Studies (EFACTS). We showed that the Scale for the
Assessment and Rating of Ataxia (SARA) is a suitable clinical rating to detect deterioration of
ataxia symptoms over time and that the activities of daily living (ADL) scale is appropriate to
monitor changes in daily self-care activities. Importantly, the rate of disease progression and
consequently the sensitivity of clinical ratings to change over time may depend on the
investigated cohort, in terms of the disease stage, age of onset or genetic burden
7-9
; and
different clinimetric properties of (sub)scales may show greater responsiveness among certain
subpopulations.
9,10
Therefore, in the present study, we describe clinical progression over four
years in FRDA and evaluate the utility of outcome measures, including subitems of SARA,
ADL and performance-based subtests of the Spinocerebellar Ataxia Functional Index (SCAFI).
By identifying different patient groups with differential progression rates over time, we further
aimed to optimize sample size calculations and eventually improve and guide upcoming clinical
trial designs.
-4-
Methods
Study design and participants
EFACTS
6,7
(www.e-facts.eu) is a prospective, longitudinal, observational cohort study based
on an ongoing and open-end registry including 11 referral centres (university hospitals and
institutes) in seven European countries (Austria, Belgium, France, Germany, Italy, Spain, UK;
table 1). Patients with genetically confirmed FRDA at a study centre were asked for interest
and participation in EFACTS. Participants, of whom we had completed and monitored four-
year follow-up data sets were included in the current analysis. All patients and/or their
authorized surrogates provided written informed consent upon enrolment into EFACTS. The
study was approved by the local ethics committees of each participating centre.
Procedures and Outcomes
Assessments were done annually in a standardized manner at each centre using the same
written study protocol. Genetic testing was repeated centrally for all patients at the Laboratoire
de Neurologie Expérimentale of the Université Libre de Bruxelles (Brussels, Belgium).
11
A
detailed description of procedures and data collection can be found in our previous reports
6,7
or online.
2,6,7
Our primary co-outcome measures were total scores of SARA and ADL, the selection was
based on our previous work
6,7
. SARA is a 40-point scale with higher scores indicating more
severe ataxia
12
, and consists of eight items pertaining to gait (score 0-8), stance (0-6), sitting
(0-4), speech disturbance (0-6), finger chase (0-4), nose-finger test (0-4), fast alternating hand
movements (0-4), heel-shin slide (0-4). The latter four items on limb kinetic functions are rated
separately for each side, and the arithmetic mean of both sides is calculated. ADL as part of
the Friedreich Ataxia Rating Scale (FARS)
13
, was carried out in a structured guided interview
setting to assess daily functional activity impairment (maximal severity-score 36). Each of the
nine items (score 0-4) measures deterioration in the respective domains of speech,
swallowing, cutting food, dressing, personal hygiene, falls, walking, sitting and bladder
function.
As secondary outcome measures, we applied the Inventory of Non-Ataxia Signs (INAS) and
calculated a count of non-ataxia signs (0-16) such as changes in reflexes, other motor, sensory
or ophthalmological signs
14
. The SCAFI consists of three timed performance-based tests
including an 8m-walk at maximum speed, the nine-hole peg test (9hpt), and the rate of
repeating the syllables “PATA” within 10 s
15
. Total composite SCAFI Z-scores were calculated
as formerly reported.
16
As a measure of health-related quality of life, we used an index of the
EQ-5D-3L.
17,18
In additional sensitivity analyses, we considered SARA, ADL and SCAFI on
subitem-level. To facilitate interpretation, we transformed SCAFI-subtest scores (i.e., m/s for
8m-walk; pegboards per minute (1/[time/60]) for 9hpt).
-5-
Statistical analysis
Data are described using mean +/- standard deviation (SD), frequencies or percentages, as
appropriate. As a measure of responsiveness of outcomes, we calculated standardised
response means (SRM), i.e. the mean change in scores from baseline to follow-up divided by
the standard deviation of the change. The yearly progression rate for each outcome was
estimated using linear mixed-effect modelling (LMEM with restricted-maximum-likelihood
estimation method) with random effects on slope (i.e., time in years [days since the baseline
visit divided by 365]) including baseline scores as fixed main effect. Additional LMEM were
used to compare progression rates in subgroups by adding interaction terms between time
and typical-onset FRDA (symptom onset at ≤24 years of age)
4
versus late-onset FRDA (≥25
years of age); or ambulatory versus non-ambulatory
7
patients. Ambulation at baseline was
defined based on a seven-item disability stage scale (spinocerebellar degeneration functional
score)
19
ranging from 1-no functional handicap but signs at examination to 7-confined to bed.
Patients being able to walk (with/without sticks, wheeled walker; score ≤5) were considered as
ambulatory, whereas patients unable to walk (≥6) were categorized as non-ambulatory. In
sensitivity analyses the problem of floor and ceiling effects, when investigating the annual
progression rate on subitem-level for SARA and ADL, were compensated by truncated
likelihood estimates (i.e. TOBIT analysis)
20
.
We further tested the effects of disease-relevant and demographic factors on progression rates
using LMEM. In addition to time, we modelled fixed main effects of study site and baseline
scores, and fixed interaction effects between time and sex, age in years at baseline,
educational level, age of symptoms onset, baseline scores of the respective outcome measure
and number of GAA-repeats on each allele. Continuous variables were mean centred to
facilitate interpretation. In order to identify cut-off values of specific disease-related factors (i.e.
GAA-repeats on allele 1, age, age of onset and baseline scores) that would enable a selection
of patients with faster disease progression, we performed breakpoint analyses of piecewise
linear regression models
21
. Finally, based on the observed LMEM progression rates (random
slopes) for SARA and ADL, we calculated total sample sizes (1:1 allocation ratio) to detect a
reduction in disease progression rates in a parallel-group interventional trial with different
treatment efficacies, visit intervals and observational periods.
Statistical analyses were done with SAS (version 9·4, procedure MIXED, NLMIXED). All tests
were two-sided with a p value of 0·05 set as the threshold for significance. The EFACTS study
is registered with https://clinicaltrials.gov (NCT02069509).
Role of the funding source
The funders of the study had no role in study design, data collection, analysis, interpretation,
or writing of the report. All authors had full access to data and took final responsibility for the