136th ENMC International Workshop: Charcot–Marie–Tooth Disease Type 1A (CMT1A)8–10 April 2005, Naarden, The Netherlands

2.1. Is CMT1A a homogeneous disorder with, at the cell biology level, a common pathomechanism that can be targeted by ascorbic acid? 

Molecular genetic studies strongly suggest that PMP22, the culprit gene in CMT1A, is a dosage-sensitive gene. The presence of three PMP22 copies, due to a tandem duplication in chromosome 17p11.2–12 leads to CMT1A while the loss of a PMP22 copy leads to a milder peripheral neuropathy, i.e. hereditary neuropathy with liability to pressure palsies. Occasionally, patients with four PMP22 copies have been reported as the offspring of couples where both parents have the CMT1A duplication and it turns out that, in general, these individuals are more severely affected. That the CMT1A duplication leads to over-expression of PMP22 at the RNA and protein levels has been well documented. All of this evidence suggests that CMT1A is a homogeneous disorder with a common pathomechanism related to gene dosage sensitivity.

2.2. Are there valid animal models of CMT1A and what have therapeutic trials in these animal models taught us? 

Over-expression of PMP22 in rodents has created animal models that closely mimic the disorder in humans. Two models have been constructed: a transgenic rat with two extra copies of the gene and pathological features reminiscent of the human disorder CMT1A [6] and a mouse strain with higher copy numbers showing a more severe, dysmyelinating phenotype reminiscent of Déjèrine-Sottas Disease or Congenital Hypomyelination [7]. How the over-expression of PMP22 causes the disease still remains to be determined but it might be that the proteosome pathway, for example by an overload mechanism and/or the formation of aggresomes, is involved [8]. Fortunately, knowledge of the disease mechanism in all its details may not be needed to develop an effective treatment. Lowering the expression of PMP22 may be all that is needed and that strategy has been addressed in two therapeutic trials in these animal models. Nave and colleagues treated their transgenic over-expressing rat model with onapristone, an anti-progesterone drug [2]. It has been known for some time that the PMP22 promoter contains steroid responsive elements and that progesterone stimulates the myelin process by enhancing the expression of PMP22. The treated rats had a lower expression of PMP22 at the RNA and protein levels and most importantly had markedly better strength and less abnormalities on nerve histopathology. Because of the known side effects of this drug its use in a slowly progressive mild disorder such as CMT1A is probably not justified. Fontes and colleagues showed that a similar effect could be achieved by the administration of ascorbic acid [3]. Ascorbic acid treatment resulted in substantial amelioration of the neuropathy phenotype of the over-expressing mice, and reduced the expression of PMP22 to a level below what is necessary to induce the disease phenotype. As ascorbic acid has already been approved by the FDA for other clinical indications, it offers an immediate therapeutic possibility for patients with the disease. However, it is important to keep in mind that the mouse model has a more severe neuropathy than seen in most patients with CMT1A due to the chromosome 17 duplication.

2.3. Is there a therapeutic time window in which the ascorbic acid would need to be given in order to be efficacious? 

This issue has been addressed by Huxley and colleagues. in a conditional PMP22 over-expressing mouse model [9]. When over-expression of PMP22 is switched off in adult mice, correction begins within 1 week and myelination is well advanced by 3 months, indicating that the Schwann cells are poised to start myelination. This suggests that ascorbic acid can be given to adults with CMT1A who already have symptoms and signs of the disease.

2.4. Will it be possible to recruit sufficiently large enough numbers of patients for a trial and do we have the tools to detect changes in the course of this slowly progressive disease? 

Since CMT1A due to the CMT1A duplication is by far the most common type of CMT, recruitment of a sufficiently large cohort of patients to give a trial enough power is not envisaged to be a problem. To date the lack of therapeutic trials in CMT has meant outcome measures have not been specifically developed to detect changes in the disease course in CMT1A patients over a short period of time. However, since CMT1A is clinically a fairly homogenous disorder with a very stereotypic progression over time, it should be possible to select outcome measures to detect altered progression or even amelioration in a trial.

3.1. Primary outcome measure 

The workshop agreed that a primary outcome measure for the trial would need to look at impairment or disability or both. In addressing these two domains, there already are single item scales and multi-quality scales (composite scales), each of which have advantages and disadvantages, in use in other neuropathy trials. The workshop did not feel that any existing impairment scale used in other neuropathy trials (e.g. MRC score in inflammatory neuropathy trials) would be suitable for the CMT1A trial because of the slowly progressive nature of the disease. Richard Hughes updated the workshop on disability scales including the overall disability sum score (ODSS) [10] and the overall neuropathy limitations scale (ONLS). He also updated the workshop on a new disability scale, the Amsterdam linear disability scale (ALDS) [11]. Most of these scales had been developed for trials of inflammatory neuropathies and it was not felt that any of these would be suitable as a primary outcome but it was agreed that a disability scale should be included as a core secondary outcome for the CMT1A trial.

Richard Lewis presented in detail the new composite scale that has been specifically developed for natural history studies and trials in CMT and which has been published in Neurology since the meeting [12]. This composite 36 point scale, called the CMT neuropathy score (CMTNS), includes measurements of symptoms, signs and neurophysiology and therefore is mainly a scale of impairment. It is well recognized that the clinimetrical aspects that need to be assessed for every outcome measure are simplicity, validity, reliability and responsiveness. The simplicity, validity and reliability of the CMTNS have already been shown but to date no studies of responsiveness are available. Richard Lewis presented some preliminary work on 45 CMT1A patients studied retrospectively using the CMTNS and reported that a 1-point change in the scale is seen over 2 years. There was much discussion as to whether a single primary outcome measure or multiple primary outcome measures should be used; the consensus was for a single primary outcome measure. There was also a consensus that a composite scale such as the CMTNS was more appropriate for CMT1A rather than a single item scale due to the slowly progressive nature of the condition but it was agreed that single item scales should be included as secondary outcome measures.

Despite the lack of responsiveness data for the CMTNS, there was a consensus that this composite score would be the most appropriate primary outcome measure for a trial of ascorbic acid in CMT1A. The workshop recognized that usually outcome measures should be completely established including robust data on responsiveness before a trial is undertaken but the workshop also agreed that there was a need for an international trial of ascorbic acid in CMT1A to be started without delay. The reason for this is that, as ascorbic acid is freely available, many CMT patients have already started to take it and Olivier Blin, Michel Fontes and Odile Dubourg reported to the workshop that they were starting a French trial of ascorbic acid in CMT1A in May 2005. The CMT patient representatives Karen Butcher and Anna Detomas agreed that the trial should be conducted soon but they also thought patients would be happy to take part in a placebo controlled trial. The final consensus was that an international trial of ascorbic acid in CMT1A should be started as soon as was feasible using the CMTNS as the primary outcome measure. For the electrophysiological section of the CMTNS, it was agreed that the ulnar nerve should be used rather then the median nerve. It was also agreed that the MRC grading should be recorded of the individual muscles as well as the actual CMTNS measurement.

3.2. Other core outcome measures 

A consensus was reached on a series of other core outcome measures. These are listed in Table 1 and include measures of impairment, of disability (activity limitation), and of Quality of Life.

Table 1. Core outcome measures

	Primary outcome measure
	Domain examined	Test/measure	Notes
	Impairment (disability)	CMTNS	Ulnar nerve to be examined. MRC of examined muscles to be recorded

	Other recommended core outcome measures
	Domain explored	Test/measure	Notes
	Impairment	Quantitative motor assessment	Hand-grip, pinch; foot movements (to be better defined)
	Disability	10m timed walking	Standardisation needed
	Disability	ONLS
	Disability	VAS	Pain and fatigue
	QoL	SF-36
	Impairment	Electrophysiology	2 motor nerves in upper limbs (ulnar and median), 1 motor nerve in lower limbs (peroneal); 1 sensory nerve (ulnar); non dominant side; MCV, DL, CMAP, SCV, SAP
		Adverse events

MCV, motor conduction velocity; DL, distal latency; CMAP, compound muscle action potential; SCV, sensory conduction velocity; SAP, sensory action potential.

It was felt that quantitative motor assessment with a myometer should be used to measure strength of handgrip and of pinch [13], [14]. The advantage is that it is a continuous and reliable measure. Quantitative motor assessment of foot movements (dorsiflexion and plantar flexion) are also important to test, but there is still a need to clearly define modalities of test, foot position, and how to overcome the difficulties of skeletal deformities, surgical corrections, tendon retractions, muscle imbalance.

Ten metre timed walking [15], [16] was considered a useful quantitative measure of disability in mobility and leg function. Standardisation is important; the test should be performed in the same place, without ankle foot orthoses (AFO) and shoes if the patient is able to accomplish the task but with support if needed, using the same support on subsequent examinations.

The workshop also agreed in recommending the ONLS as the core disability scale and the SF-36 questionnaire [17] as the core QoL scale.

It was also agreed that a visual analogue scale (VAS) to evaluate pain and fatigue should also be included among outcome measures.

The recommended core electrophysiological evaluation protocol comprises the evaluation of two motor nerves in upper limbs, one motor nerve (peroneal nerve) in lower limbs, and one sensory nerve in upper limbs. The protocol includes the motor and sensory ulnar nerve, which is also examined as part of the CMTNS. The non-dominant side should be tested, temperature carefully checked and motor conduction velocities (MCV), compound muscle action potential (CMAP) amplitudes, distal latencies (DL), sensory conduction velocity (SCV) and sensory action potential (SAP) amplitude measured.

The workshop agreed that in any trial adverse events must be a core outcome measure.

Written instructions for all tests and outcome measures should be prepared.

The use of other more complex disability scales can be considered (i.e. Amsterdam linear disability scale, [11]) but they were not included in the core outcome measures.

The use of new techniques to study outcomes may prove valuable for CMT trials and their use should be encouraged. These techniques could be incorporated into a trial protocol and their usefulness in CMT trials could be investigated in this way. An example of this is the use of skin biopsies to measure the expression of PMP22 at the mRNA or protein level, since it can provide a biological marker of treatment efficacy [18]. Josè Berciano showed that magnetic resonance imaging (MRI) of muscles may detect pre-symptomatic changes and be a useful early marker of disease [19]. MRI of nerves is also a promising technique [20]. Therefore, additional outcome measures may be studied by some groups and are to be encouraged.