Selecting Outcome measures for MITOchondrial disease in children (the SO-MITO study)

In recent years, there is more and more interest in the treatment of
mitochondrial dysfunction by e.g. antioxidants1-4. However, the quality of
these studies is widely criticized, since most studies are of inferior
scientific quality and have insufficient sample size5, 6. One of the aspects of
scientifically sound research in mitochondrial disorders we are currently
focusing on is the selection of clinically relevant and scientifically robust
outcome measures for children.

Our approach and past research
In Figure 1 you can see our approach. After an inventory of the most burdensome
symptoms rated by patients and their parents7, we*ve reviewed the literature
for available instruments which were previously able to measure these aspects
reliably in disorders with similar symptoms and limitations8. In this review,
we*ve presented a toolbox of 33 instruments which, in our opinion, are the most
promising instruments to measure disease severity and disease progression in
children with mitochondrial disorders. However, there is no experience with
these instruments in children with mitochondrial disorders yet.

The challenge of selecting outcome measures for mitochondrial disease
Internationally, there is a huge demand for so-called *common data elements*
for diseases like mitochondrial disease (http://www.nlm.nih.gov/cde/) . This is
an extremely challenging task, since mitochondrial disorders are extremely
heterogeneous. Since mitochondria are present in virtually every cell of the
human body, virtually every organ can suffer from mitochondrial dysfunction.
Mostly, mainly the most energy-consuming organs are affected, including the
brain, muscle, heart and eye in every conceivable combination and severity. For
example, a patient with Leigh syndrome, the prototype paediatric mitochondrial
disease, may suffer from severe psychomotor retardation, dystonia, brainstem
failure and responds poorly to his environment, whereas a patient with
Kearns-Sayre syndrome suffers from chronic progressive external
ophthalmoplegia, muscle weakness, retinopathy, but may for example be able to
attend school. From a molecular perspective, more than a hundred genetic
defects causing mitochondrial disorders are known (and many more are to be
found using whole exome sequencing), causing a wide variety of biochemical
defects with their corresponding cellbiological consequences9. From a clinical
perspective, there is a very poor genotype-phenotype correlation, even between
siblings10. Moreover, the traditionally used and well-defined syndromes (Leigh
syndrome, MELAS syndrome, etc) may have many different underlying molecular
mechanisms.

Patient inclusion
One of the many questions that has to be answered before we can recommend
certain outcome measures to be used in clinical trials in children with
mitochondrial disease, is if we should select patients for intervention studies
based on biochemical or phenotypical characteristics. Please see Figure 2 for a
summary of the pro*s and con*s of both approaches. In this study, we aim to use
both perspectives to group patients. For the clinical syndromes, we chose to
include the well-characterized mitochondrial syndromes, such as Leigh syndrome,
MELAS syndrome, Kearns-Sayre syndrome and Barth - and Sengers syndrome. The
patients in the other group will be included based on their biochemical data
and subsequently be classified into three clinical subgroups: myopathic and
encephalo(myo)pathy patients, based on strict criteria.

Primary Objective:
*Which outcome measure should we use to measure disease severity and disease
progression in this specific group?*
To test the feasibility (% of patients who were able to complete the test) of
the selected outcome measures.
To test the reliability (inter-rater reliability, intra-rater reliability,
test-retest reliability) of the selected outcome measures.
To test the validity (correlation with predefined anchors and between
parameters measuring the same construct) of the selected outcome measures.
To test the responsiveness over time of the selected outcome measures.

Secondary Objectives:
*Is it feasible to use this group for intervention studies?*
To determine the SD and the minimal important difference by the patient for the
two most reliable outcome measures and determine the sample size for future
studies
To determine the time to complete the patient inclusion

*What are the functional abilities and challenges of the specific groups (which
should be given more attention to in clinical practice)?*
To determine the clinical spectrum of the phenotypes, with a specific focus on
the function abilities and challenges

*What is the value of the commonly used biomarkers for mitochondrial
dysfunction as an outcome measure in intervention studies?*
To determine the correlation between the commonly used biomarkers for
mitochondrial dysfunction and the functional capacities and disease progression
in this group

This study is an observational study, in which the observation is performed
using several instruments which were previously selected to be promising tools
to measure disease severity and disease progression in children with
mitochondrial disorders 8.

Patient inclusion and protocol selection
Patients will be included from two perspectives: the syndromal and the
biochemical approach. The patients with a well-characterized syndrome (see
paragraph 4.2) will be included in the syndromal group. Patients with a
well-characterized mitochondrial dysfunction in combination with a classical
mitochondrial phenotype (see paragraph 4.2) will be included in the biochemical
group. The patients within the biochemical group are classified into the
myopathic and the encephalopathic group (the clinical subgroups) based on their
clinical symptoms. For the patients within the syndromal group, the syndrome is
classified into a specific clinical subgroup, so all patients with the same
syndrome perform the same tests. Each of the two clinical subgroups has its own
toolbox of instruments measuring the functional abilities of the patient. For
each syndrome, a number of instruments will be added (syndrome specific
toolbox) based on the syndrome specific complaints and symptoms in this group.
See also paragraph 5 for the methods and a description of the toolboxes.

Study design
These instruments will be used a time x, time x + 2 weeks and time x + 9
months, see Figure *. The patients will be seen by two experienced assessor
(paediatric pediatric physiotherapists), who will both keep their own CRF. The
examinations will be videotaped and, where possible, be scored again after 4
weeks to determine inter-rater reliability. Test-retest reliability will be
tested by measuring at time x and at time x + 2 weeks. Patients will be phoned
two days prior to the second assessment to ask if the disease has been stable
since the previous assessment. If not, they will not be seen at time x + 2
weeks, but at x + 9 months and at x + 9.5 months. Responsivity will be tested
by seeing the patients again at x + 9 months. At all visits, patients, parents,
assessors and the physician are asked how they estimate certain aspects of the
disease severity of the patient, see attachment *.

Patients will be seen at the outpatient clinic of the assessors three times.
This means that they will not be able to attend school during that day. Since
these patients have a mitochondrial disorders and tired easily, it is possible
that they will be more tired during the week. Also, it is known that patients
with mitochondrial disease may suffer from muscle aches after exercise. Since
we will test the limits of the patients* possibilities, this will probably also
be the case after our examinations. The risks of serious complication of the
individual tests are however, in literature and in our experience, negligible.
We have deliberately chosen for instruments measuring the abilities in daily
life and not for strenuous test such as the maximal exercise test, because
these tests are probably too burdensome to the patients. The burden will also
be minimized by a balanced programme, including regular pauses and alternation
between exercise tests and more quiet tests, see attachment *.

On every visit to the outpatient clinic, after all tests are performed, blood
will be drawn for biomarker analysis. We will try to make this as comfortable
as possible (e.g. using emla cream), but it may still be a burden for the
children. We draw blood at the end of the day because of this may be a
traumatic experience for the child.

For parents, there will also be a burden, by taking 3 days off. Since we give a
detailed programme including information on all three days, we think that
parents may be able to make a decision for themselves.

Since we aim to test paediatric measurement instruments, this study cannot be
performed in adults. We think this study is relevant to all children with
mitochondrial disease, since studying future compounds in a homogenous,
well-characterized cohort of patients using feasible, reliable and valid
instruments increases the likelihood that valid conclusions will be drawn
according to the effect of the treatment. Also, this study is an expression of
our aim to advocate scientifically sound research in orphan diseases, which is
sadly still rare.

This study will give a good overview of the patient*s health state and it might
give a better overview over the patient*s problems than can be obtained in a
regular outpatient visit. Therefore, a summary letter with the findings of this
study will be sent to all physicians caring for the patient. This may be
beneficial for the coordination of the care around the patient.