Neuromuscular Disorders
Volume 12, Issue 2 , Pages 201-210, February 2002

90th ENMC International Workshop: European Spinal Muscular Atrophy Randomised Trial (EuroSMART) 9–10 February 2001, Naarden, The Netherlands

  • Luciano Merlini

      Affiliations

    • Neuromuscular Unit, Istituto Ortopedico Rizzoli, Via Pupilli 1, 40136 Bologna, Italy
    • Corresponding Author InformationCorresponding author. Tel.: +39-051-332525; fax: +39-051-332920
    • ,
  • Brigitte Estournet-Mathiaud

      Affiliations

    • Hôpital Raymond Poincaré, Service de Pédiatrie, Garches, France
    • ,
  • Susan Iannaccone

      Affiliations

    • Department of Neuromuscular Diseases and Rehabilitation, Texas Scottish Rite Hospital for Children, Dallas, TX, USA
    • ,
  • Judith Melki

      Affiliations

    • Laboratoire de Neurogenetique Moleculaire, Genopole, INSERM E. 9913, Evry Cedex, France
    • ,
  • Francesco Muntoni

      Affiliations

    • Department of Paediatrics and Neonatal Medicine, Hammersmith Hospital, Imperial College School of Medicine, London, UK
    • ,
  • Sabine Rudnik-Schöneborn

      Affiliations

    • Institute for Human Genetics, University Hospital, RWTH Aachen, Germany
    • ,
  • Haluk Topaloǧlu

      Affiliations

    • Department of Child Neurology, Hacettepe Children's Hospital, Ankara, Turkey
    • ,
  • Giuseppe Vita

      Affiliations

    • Clinica Neurologica 2, University of Messina, Messina, Italy
    • ,
  • Thomas Voit

      Affiliations

    • Department of Pediatrics, University Hospital Essen, Essen, Germany

Article Outline

 

Back to Article Outline

1. Introduction 

Spinal muscular atrophy (SMA) is the most common form of motor neuron disease in children and young adults with an incidence up to 11 per 100,000 live births [1]. The disease is characterized by degeneration of anterior horn cells leading to progressive paralysis with muscular atrophy. Depending on the clinical type (Werdnig–Hoffmann=type I, intermediate form=type II, Kugelberg–Welander=type III) [2], SMA causes early death or increasing disability in childhood or adulthood. The underlying genetic defect on chromosome 5q13 has been identified and characterized [3]. This chromosomal region contains a duplicate survival motor neuron gene (SMN). One copy of SMN (termed telomeric SMN or SMN1) shows homozygous mutations or deletions in SMA patients, whereas the second copy (termed centromeric SMN or SMN2) is not affected.

The full-length protein is almost exclusively produced from SMN1. In contrast, the primary transcript of SMN2 undergoes alternative splicing of exon 7, which results in the predominant expression of an unstable and C-terminally truncated SMN protein [3]. Although the SMN2 gene remains present in patients, the protein derived from SMN2 is clearly insufficient to prevent motor neuron damage. At the cellular level, this leads to apoptotic cell death of motor neurons in the anterior horns of the spinal cord and consequently to SMA [4]. In SMA patients, full-length SMN protein levels and the number of gems generally correlate with disease severity, suggesting a critical nuclear function for SMN [5]. At the moment, the function of the protein encoded by the SMN gene is still not fully understood, but recent studies suggest involvement of the SMN protein in the formation of spliceosomal particles in the cytoplasm and in the regeneration of spliceosomes in the nucleus. Its interaction with other proteins suggests a role in nuclear pre-mRNA splicing and transcription regulation. Despite recent progress in the understanding of the cellular function of SMN, the link between SMN defect and SMA pathophysiology remains to be established. In particular, it is unclear why reduced levels of functional SMN protein can lead to specific dysfunction of motor neurons without affecting other cell types and tissues.

Multicenter randomised clinical trials in SMA have not been published yet. However, after the 38th ENMC workshop devoted to the formation of a spinal muscular atrophy trial group [6], two multicenter trials have been promoted. An 8-city gabapentin drug trial (adults aged 18-60) began in March 1998 in the USA under the coordination of Robert G. Miller on behalf of the SMA Study Group. Italian Telethon recently granted a Spinal Muscular Atrophy Randomised Trial (Italian SMART) with gabapentin (120 patients aged 5–60) coordinated by Luciano Merlini (Bologna) and with the participation of six other investigators/centres: T. Mongini (Torino), L. Morandi (Milan), C. Minetti (Genova), C. Angelini (Padova), E. Bertini (Roma), G. Vita (Messina).

Few pilot studies of SMA trials have been published yet. An anabolic steroid and hexahydrocoenzyme Q4 were tried in a case of spinal muscular atrophy [7]. A double blind controlled study with guanidine hydrochloride in infantile and juvenile spinal muscular atrophy in few patients showed no clear-cut effect [8]. The most recent one is a randomised double-blind controlled 5-wk drug trial of six subjects and three controls. TRH (protirelin) or placebo was delivered intravenously through percutaneous intravenous catheters at a dose of 0.1 mg/kg (in 50 ml of normal saline) for a total of 29 days. Dynamometry improved significantly only for the six treated subjects (P=0.02). Peroneal nerve conduction velocities were significantly faster in the treatment group (paired t-test, P=0.036). The parents of the treated children also provided anecdotal evidence of improvements in function. Improvements lasted 6-12 mo. The authors concluded that TRH might be a useful treatment for spinal muscular atrophy [9].

Back to Article Outline

2. Clinical trial experience 

Guiseppe Vita (Italy) presented the Italian Spinal Muscular Atrophy Randomised Trial (Italian SMART), sponsored by the Telethon Foundation. It is a seven-centre randomised open comparative parallel-group trial of gabapentin in SMA. Gabapentin was identified as a potential useful drug for different reasons: (i) Glutamate-induced neurotoxicity is a common pathogenetic event for both acute and chronic neuronal death, and although direct evidence for glutamate-induced cellular damage in SMA is lacking, some experimental data suggest that such indeed may be the case [10]; (ii) Glutamate inhibitors, such as riluzole and gabapentin, delay degeneration of motoneurons and prolong survival in a transgenic mouse model of ALS (mSOD1) [11]; (iii) there are some clinical and pathogenetic similarities between SMA and ALS. SMART started in March 2000, with the primary aim of testing the efficacy of gabapentin in improving muscle strength and function of patients with type II and type III SMA. The secondary aims were (i) to better define the natural history of the disease, and (ii) to develop valid, reliable, and responsive measures to assess the outcome of clinical and experimental studies on SMA patients. One hundred and twenty patients aged 5–60 years, and carrying SMN gene mutation, have been recruited. Drug dosage was titrated up to 50 mg/kg or up to a maximum of 1800 mg in 6 weeks (starting dose: 1/6 of the maximum). The duration of the trial was twelve months. The evaluation tests included arm myometry (elbow flexion, pinch, grip and arm megascore as the sum of three previous values), leg myometry (knee extension, knee flexion, foot dorsiflexion, and leg megascore), forced vital capacity (FVC) on spirometry, and timed tests (walking 10 m, climbing and going down three standard steps, rising from the floor). The primary outcome target was a 35% increase in baseline strength of the arm megascore; the secondary outcome targets were (i) a 35% increase in baseline strength of the leg megascore and (ii) a 35% increase in baseline FVC. Vita mentioned the recent results of a US double-blinded placebo-controlled study of gabapentin (1200 mg/die) in adult (18-60 years) patients with SMA, which were presented by Miller et al. at the Annual Meeting of the American Neurological Association in Boston, Massachusetts on October 17, 2000 [12]. Forty patients were randomised to receive gabapentin and 44 to receive placebo. Sixty-four patients (76%) completed the twelve-month study. The mean rate of change in arm strength (a composite of four arm muscles), the primary outcome measure for that study, was not statistically significantly different for patients receiving gabapentin compared with those receiving placebo. Moreover, there was no significant difference between patients treated with gabapentin and placebo in the following secondary outcome measures: rate of strength change in those completing the entire study, rate of change in respiratory muscle function, rate of change of function on scales, and change in quality of life. Gabapentin was generally well tolerated, with no relevant side effects more frequent in those receiving the drug. Miller concluded that there is no evidence that gabapentin has any disease-specific beneficial effect upon muscle strength, respiratory function, or overall function of patients with SMA in this controlled clinical trial. Because of the difference in patient population age and drug dosage, SMART investigators, together with Telethon, decided to continue the Italian trial. The Italian Smart will be completed in June 2001.

Francesco Muntoni (UK) reported the first results of a pilot trial of salbutamol in 13 SMA patients from Hammersmith Hospital, London. Seven were SMA2 and six SMA3; age range was 5–21 years; mean age: 13.1 years. Patients received salbutamol at a dosage commonly used for asthma (2 mg tid/qid) in oral solution or tablets. Measures of efficacy were the change from baseline to 12 and 24 weeks in muscle strength, assessed by MRC score (qualitative evaluation of 32 muscles) and myometry (16 muscle groups), in FVC, functional scores, and in lean body mass by using DEXA scanning. Statistical analysis was performed using repeated measures ANOVA (significance level P<0.05). Tolerance of the medication was good in all patients. Overall there were encouraging results; there was a significant increase in myometry scores and FVC between the baseline and the 24-week assessments (P<0.05). The lean body mass also significantly increased between baseline and 24 weeks (P<0.05), while the increase in MRC scoring reached significance after 12 weeks but not after 24. Three of the patients with SMA III reported a significant improvement in function (fewer falls, longer walks). The functional score in the patients as a whole, however, did not change significantly from baseline.

Haluk Topaloǧlu (Turkey) reminded the group that carnitine is a transporter for long chain fatty acid into the mitochondrial matrix, and that denervation leads to a secondary defect in muscle intramitochondrial beta-oxidation [13].

In addition, it has been shown that muscle carnitine is reduced in SMA [14] and that reduction is related to the severity of denervation. Secondary metabolic defects have been demonstrated in spinal muscular atrophy type II [15], including muscle carnitine decrease.

In his trial, Topaloǧlu included 23 SMA1 and 14 SMA2 patients; carnitine was given at the dosage of 100 mg/kg/day. The primary outcome was the reduction in frequency of pulmonary infections and mother's impression of well-being. Surrogate markers were body weight, triceps skin fold, mid-arm circumference, and daily diet intake.

Although there was no variation in the age of death, treated patients suffered fewer episodes of pulmonary infections, had decreased vomiting, and an increase in growth centiles. The mother's impression was also favourable.

Thomas Voit (Germany) reported on an ongoing 6-month trial conducted at the University of Essen, Germany with alpha glucosidase in two patients with the infantile subtype of glycogen storage disease type II (Pompe disease). Inclusion criteria were: age under 6 months and presence of cardiomyopathy in addition to skeletal myopathy and low activity of alpha glucosidase. The first patient was terminal at the start and on continuous oxygen support. Four months after treatment he was on intermittent oxygen support and tube feedings, with pulmonary oedema and right ventricular failure. The second child was a girl who was continuing to progress; she was able to sit upright for a few seconds and roll over. Adverse events included allergic reactions to the medication evidenced by tachycardia, vomiting, and fever. Echocardiographic evaluations showed a regression of the left ventricular posterior wall thickness and the left ventricular mass index in both patients. Treatment consisted of 40 mg/kg of alpha glucosidase by intravenous route once weekly, obtained from the milk of transgenic rabbits (Pharming, Genzyme). The major outcome of this open label study was prolongation of survival.

Susan Iannaccone (USA) presented the results of the DCN/SMA group (DCN for Dallas-Cincinnati-Newington), which initially focused its interest on the natural history of the disease [16], [17]. This study was undertaken in order to define the natural history of SMA and to determine whether patients suffer progressive loss of strength. Patients were monitored biannually for muscle strength, respiratory function, and survival up to eight years. No patient was on ventilatory support. A custom designed table was constructed for each of the three centres after it was determined that positioning of children for quantitative muscle test (QMT) was problematic. A total of 159 SMA patients were enrolled in the study. Each evaluation session included QMT for the following muscle groups: hand grasp, elbow flexion, knee flexion, and knee extension. The sum of this formed the total muscle score. Functional categories were three: non-sitter, sitter (sit without support when placed), and walker (walks independently). For purposes of this report, a change in function was defined as moving from one function group to another during the study. They showed that fifty percent of SMA patients who could walk without assistance and whose onset was prior to age 2 lost the ability to walk independently by age 12. Fifty percent of SMA patients who walked and whose onset was between 2 and 6 years of age lost walking ability by age 44. Fifty percent of SMA patients who could walk with assistance as their best function ever achieved lost this ability by age 7, unrelated to age of onset; none could walk with assistance after age 14 [17]. They concluded that all patients with SMA lose function over time. This functional loss occurs slowly and is related primarily to maximum function achieved; knowledge of age of onset provides helpful information, especially for predicting the loss of independent walking [17]. There was a definite difference between changes for patients depending on the age group. For children under age 15, the SMA subjects had smaller lung capacities. After age 15, there was a larger difference in FVC between SMA (mean, 1.8 l) and normal individuals (mean, 2.8 l). Likewise, the total muscle score (TMS) was significantly lower in SMA patients than in controls in both age groups. For young SMA subjects, the mean TMS was 17.1 kg; for 15 years and older, it was 21.1 kg. Normal subjects had TMS of 101.7 and 229.5 kg respectively. The difference between younger and older SMA patients was only about 25% (17.1 to 21.2 kg) as compared to under- to over-15 normal subjects, which was 200%. A total of 10 patients lost function; no one gained function. For children under age 15, whether they lost function during the study did not affect the change in weight, height, FVC, or TMS. The increment in TMS was 4.2 kg for subjects who did not lose function and 3.4 kg for subjects who lost function. They concluded that muscle strength as measured by QMT did not decrease during an observation period of 2–6 years in SMA. The increase noted for patients under age 15 was statistically significant. Such improvement during early childhood may be attributed to growth and development, but remains far below what is expected for normal children. During the DCN study, it was possible to document loss of motor function such as ambulation or sitting ability, despite the lack of loss of quantitative muscle strength.

Sabine Rudnick-Schöneborn (Germany) summarised the conclusion of a German study aimed to evaluate the attitudes toward future trials and spinal fusion. The inquiry involved 169 families (105 patients and 64 parents) with chronic proximal SMA of all types of severity. A considerable fraction of patients (20–30%) had principal objections against clinical trials, mainly because their families coped well with the situation and were therefore critical regarding potential risks of future treatment. Parents and patients of chronic SMA families expressed similar attitudes, although parents were less likely to tolerate adverse effects or potential deterioration of a treatment in comparison to patients. There was consensus that clinical trials are not acceptable in early infancy or childhood. In addition, it was found that the drug delivery had an important impact on the attitude towards clinical trials, i.e. daily injections were regarded as too invasive for many families. It was interesting to see that somatic gene therapy was not considered substantially different from conventional drug treatment in this well-informed group of patients.

Back to Article Outline

3. Clinical evaluation of SMA patients 

It has been shown that all patients with SMA lose function over time [17] and that loss of function is due to muscle fibre loss [18]. In fact, the presence of a segmental distribution with preservation of good strength in some of the muscle in patients at an early stage of the disease was followed with the virtual absence of a segmental distribution with more widespread weakness in those at a late stage, suggesting that the mechanism of progression is loss of muscle fibres [18]. The same mechanism was proposed through morphological and electrophysiological data [19].

These data are relevant for clinical trials in SMA in several ways. Clinical and laboratory data support the concept that in SMA the progressive loss of function is due to muscle fibre loss by further motor neuron loss. Neurotrophic factors or neuroactive drugs could prevent further motor neuron and motor function loss. In addition quantitative muscle strength evaluation may be selected as a useful surrogate end point. In fact, it is non-invasive as compared to electrophysiological testing, and may reveal small changes in muscle strength that are not large enough to cause muscle function improvement or deterioration.

Brigitte Estournet (France) reported on a French 2-year prospective multicenter study on clinical evolution of SMA. A total of 236 cases were included: 68 type I, 136 type II, and 32 type III from 11 centres. Collection of data was made by a physician on a questionnaire, which included motor function, ability to perform voluntary movements, activities of daily living, angular deformity of the spine, respiratory function, and digestive tract complaints. They studied in each type of SMA some indexes and their evolution 2 years after on 62 patients. Motor functions index (MFI) is the sum of the various functions, scored as 0 (no function), 0.5 (partial ability), and 1 (normal function), divided by the number of subjects in each group. The functions include: head control, independent sitting, sitting unassisted, independent standing, able to stand unsupported, crawling, independent walking, supported walking, and walking with orthosis. MIF resulted 0 in type I, 1.4 in type II, and 6.2 in type III. The MIF evolution at 2 years on 48 cases was: 1.4 to 1.21 (P=0.08, matched pairs). The physical examination movement index (PEMI) is the sum of scores (0, 0.5, or 1) for forearm raising, arm raising, getting an object to mouth from supine position, getting an object to mouth from sitting position, leg raising with thigh supported, lower limb raising, head raising from prone, head raising from supine, turning around alone, movements of fingers and hand, and movements of foot. PEMI at baseline was 5,3 in type II, and after 2 years was significantly reduced at 4.26. Vital capacity decreased, also in a significant way, from 50% to 44% in 68 patients in 2 years. In conclusion, worsening was observed in almost all the cases at any age with a 2-year survey. If the completed study produces the same results, a multicenter therapeutic trial could be done on types I to III, regardless of age, with a 2-year survey, and this clinical protocol could be used for evaluation criteria.

Luciano Merlini (Italy) reported on the baseline evaluation of 120 SMA patients who were recruited for the Italian SMART. There were 77 (43 males–34 females; mean age 22.7 years) non-ambulant and 43 (25 males–18 females; mean age 25.1 years) ambulant patients. In the two groups, sex and age were not significantly different. Hand-held dynamometry was significantly different (P<0.0001) in the two groups for all tests (arm-megascore, leg-megascore, grip, pinch, elbow flexion, knee extension, knee flexion, and ankle dorsiflexion). Thus, they were able to show that in a large number of SMA patients, muscle strength measured with hand-held myometry highly correlates, both in the upper and lower limbs, with muscle function, i.e. the ability to walk. In addition, muscle strength in the six tests was very limited in SMA patients compared to the normal population. For example mean knee extension was 5 N (Newton) in non-ambulant and 20 N in ambulant patients. The 50th centiles value of 50 normal females and males (average of three measurements, left and right combined) was more than 160 N for knee extension, and for knee flexors, 122 N in female and 162 in male [20]. The value of 23 adult SMA patients able to walk was 11 N for knee extension and 34 N for knee flexion. The mean value of elbow flexion was 19 N in non-ambulant and 82 N in ambulant patients. A similar ratio of four was evident in most of the measures, if considered as a mean value in non-ambulant versus ambulant patients (pinch 8 N versus 35 N; knee extension 5 N versus 20 N; knee flexion 14 N versus 50 N; ankle dorsiflexion 10 N versus 42 N). In the ambulant patients, mean knee extensor strength was 12 N in the 23 adults (aged 18–55) and 29.7 N in the 20 children (aged 5–17). This difference is very significant (P=0.001) and is compatible with the downhill course of the disease. Note also that all normal subjects are stronger than 160 N, which is the maximum value possible to be measured in the test position for knee extensors. Also, the non-ambulant patients showed a significant difference in knee extensor strength (P=0.005) with age; the 48 adults had a mean of 4 N versus 7.2 N of the 29 children. In conclusion, the baseline evaluation of the 120 SMA patients enrolled in the Italian SMART showed that QMT with hand-held myometry is feasible in a wide range of patients, and data correlates with residual functional ability and age.

Francesco Muntoni (UK) reported the results of the assessment of a group of 60 children with SMA using a novel functional motor scale developed at Hammersmith. This was devised for use in children with spinal muscular atrophy, especially those with the intermediate form, to give objective information on motor ability and clinical progression. The scale, which has 20 scored activities, was designed to be self-explanatory, quick, easy to use, reproducible, and reliable. An accurate indicator of functional ability can be obtained from children as young as 12 months, up to and including all the activities achieved by the age of 3 years in children with normal developmental milestones. This scale is now being routinely used in SMA patients who are wheelchair-dependent and could be used as a tool to monitor disease progression and response to treatment.

Back to Article Outline

4. Standard scientific procedures 

Methodological errors abound in medical research, and prominent among them is the inappropriate use of statistical methods [21]. The concept of the relation between measurement scales and statistical methods is attributed to Stevens [21], who identified 4 scales of ascending levels (nominal, ordinal, interval, and ratio). In the nominal scale, observations are in mutually exclusive and exhaustive classes in which order, interval, and ratio are meaningless. Medical examples are alive/dead, ambulant/non-ambulant. The ordinal or ranking scale also has mutually exclusive and exhaustive classes, but there is order between them, as indicated by expression such as ‘better’ or ‘more’. Rating scales of all kinds and the MRC scale of muscle strength evaluation are examples of ordinal scales. Interval and ratio scales are continuous, have equal intervals, and have measurement unit such as: cm, g, °C. On interval scales, the difference between values have the same meaning whatever the size of the measurements, whereas on ratio scales the ratio of the two values has a constant meaning. A survey of 12 medical journals showed that in at least 75% of 175 papers employing ordinal measurements scales, statistical methods were used, which do, in fact, assume a more refined measurements scale [21]. One of the aims of this workshop was to fully appreciate the significance of classes used particularly in muscle strength and function evaluation in order to apply the proper statistics.

The manual muscle test devised by Kendall and by the Medical Research Council is very reproducible in well-trained hands; however, the results of the cooperative Duchenne muscular dystrophy study group demonstrated that there might be a tremendous variation between clinicians [22]. In the MMT, the values, whether normal, good, fair, or 5, 4, 3 are descriptor of strength, not absolute measures [23]. Thus, they are ordinal numbers; only the order of the numbers is meaningful, whereas the distance between two grades does not lend itself to practical interpretation [24]. The significance of this in an isolated clinical setting may be minimal, however, for determination of clinical trials it is of great importance. In addition, the appropriate statistics for significance of ordinal data are non-parametric tests, which are less powerful than the parametric methods that can be used with interval or ratio values. Since outcome measures are an important issue in rehabilitation, objective measurements of strength should be used in clinical settings. Considering cost and assessment time, the myometry technique seems to be highly valuable [25]. In addition, myometry measurements are more sensitive to detecting increases in strength over time, which are not reflected by changes in MMT scores [26].

Giovanni Savettieri (Italy) summarised the characteristics of the various types of trials. The final aim of a clinical trial is to determine whether a given treatment is beneficial among subjects under strictly controlled conditions. A clinical trial is an experiment that can be made following different strategies. The correct strategy depends on the drug one is going to study, the natural history of the disease under investigation, and some ethics concerning treatment of the disease. Here, the structure of three different clinical trial designs will be briefly described. The parallel group design (between group-comparison) type of clinical trial compares the results of treatment on two separate groups of patients. In this case, the participants are randomly assigned to one of two groups (i.e. experimental group or placebo group). The outcome values of each group are then compared. The factorial design type of clinical trial evaluates two or more treatments with a control group. Individuals constituting the sample are randomly assigned to the arms of the study. The outcome values of each group are then compared with each other and with the placebo group. A crossover study compares the results of two treatments on the same group of patients. Patients are randomised to a sequence of treatments. In this type of trial, patients serve as their own control. The major advantage of this study design is that the sample size can be reduced. Disadvantages are related to order effects and treatment carry-over. Randomisation is a critical step for each type of clinical trial. Simple randomisation randomly assigns patients to either a treatment or control group. In the case of blocked randomisation, randomisation is set up within ‘blocks’ of equal size. Stratifying within subgroups defined by prognostic factors characterize the stratified randomisation. Finally, a specific type of randomisation is adaptive randomisation, in which randomisation changes as the trial progresses to achieve simultaneous balance with several factors. Endpoints can be defined as measures of treatment response. They must be valid and reliable, but also sensitive to clinical changes. Finally, a well-designed clinical trial has to be carefully controlled for biases due to errors in the sample selection or in data collection.

Massimo Musicco (Italy) pointed out that SMA classification is mainly based on severity of presentation and progression of symptoms and signs (type I, II, and III); however the three types represent a continuum of severity. Also, familial cases may present with varying severity. This heterogeneity has consequence for clinical trials in that large number of patients is needed, since non-responders are inevitably included and outcome occurrence cannot be precisely defined. This means that a multicenter approach is needed, with consequent organizational difficulties and the need of precise standardization of procedures. In any clinical trial it is important to evaluate the advantages of the studied intervention on clinically relevant end-points. However, in SMA, some clinically relevant end points such as death, ability to walk, and autonomy in daily life activities require long-term observation. For these reasons, adequate SMA trials are expected to be large and of long duration. One possible solution to these difficulties is to use surrogate end points. Surrogate end points are more sensitive to changes induced by treatments and can reduce the observation time and/or the number of patients needed, and are generally treated as continuous instead of categorical variables. However, surrogate end-points are often unreliable as long-term prognostic predictors. One frequently used surrogate end-point is the preservation or increase of muscular strength. More information is needed about muscular strength in clinical trials on SMA. Is muscle measure reliable and reproducible? Is muscular strength a valid predictor of patient prognosis? Another important point is related to randomisation. Is it feasible in patients with SMA? What is the position of the parent/patient support group about this item? Another point to be considered is if it would be reasonable to gather more information about the natural history of SMA before starting with formal large randomised clinical trials.

Back to Article Outline

5. Reliability of outcome measures in SMA 

Luciano Merlini (Italy) reported on the reliability of hand held myometry in patients with SMA. In the Italian multicenter trial on the efficacy of gabapentin in spinal muscular atrophy (SMA) type II–III, the main outcome measure was maximal voluntary isometric contraction (MVIC) measured in upper and lower limb muscle groups by hand-held myometry. Preliminary assessment of reliability of MVIC measurement between and within raters was part of the trial. MVIC was evaluated by hand-held myometry in 33 SMA patients (17 males), with a median age of 22.9 years (SD 11.4, range 5-64 years). The following movements were examined: elbow flexion, handgrip, three-point pinch, knee flexion, knee extension, and foot extension. Each movement was assessed on the patient's preferred side only. The patient performed each movement three times for 3–5 s, with 6–12 s of rest; the highest score obtained was considered. Each patient was independently tested by two evaluators, and re-tested by one of the two evaluators, with at least 20 min rest between tests. Inter-rater testing order was randomly determined. Reliability was assessed by means of the intraclass correlation coefficient (ICC). Inter-rater reliability was good to excellent for upper limb strength, with ICC ranging from 0.92 for three-point pinch (95% confidence interval [CI] 0.86–0.95) to 0.98 for elbow flexion (95% CI 0.96–0.99). In lower limbs, reliability was moderately good to excellent, with ICC=0.70 for foot extension (95% CI 0.47–0.84), and 0.95 for knee flexion (95% CI 0.90–0.97). Intra-rater results were excellent with ICC >0.91 in all instances. Merlini concluded that MVIC measured by hand-held myometry is easily performed in SMA patients at different ages and degrees of strength. It is a reliable measure of limb muscle strength and can be implemented in longitudinal studies and clinical trials.

Back to Article Outline

6. American SMART 

Susan Iannaccone (USA) presented the recently NIH funded American Spinal Muscular Atrophy Randomised Trial (AmSMART) project. The first goal of this project was to organize a multicenter clinical investigation in children with SMA. The AmSMART group is an organization of five paediatric medical centres formed in order to perform clinical trials in children with SMA. The Coordinating Centre is Dallas, where the Principal Investigator, the Project Coordinator, two physical therapist evaluators, the pharmacy and the Data Management and Statistics Centre are located. The Grants Management Office of UT Southwestern Medical Centre (UTSW) administers the grant, while patient enrolment occurs at the five paediatric centres with which UTSW has consortium agreements. Each participating centre other than Dallas has a paediatric neurologist as sub-investigator, a physical therapist evaluator, and a nurse coordinator. Each paediatric centre has been supplied with the same equipment. The second goal was to devise reliable methods to measure strength, motor function, lung function, and quality of life, all of which may be used as outcome measures in children with SMA [27]. The development of objective and quantitative parameters for evaluation of patients will be necessary for judging therapeutic effectiveness. The QMT will be used for measuring muscle strength. The Gross Motor Function Measure will be used for motor function assessment. Lung function will be measured using forced vital capacity, maximum inspiratory and expiratory pressures, and peak cough flow. Quality of life will be evaluated with the Paediatric Quality of Life Instrument. Each method will be tested for reliability in SMA children and modified as needed to provide accurate results in this population. The final goal was to determine whether administration of creatine to children with SMA is safe, and whether it might improve muscle strength, motor function, or lung volumes. Creatine has been shown to improve strength in a transgenic animal model of motor neuron disease as well as in patients with motor neuron disease. Creatine will be given in a placebo-controlled blind trial. This pilot project will be an important step toward the ultimate goal of finding an effective treatment for SMA by offering a group of highly experienced investigators an opportunity to refine methods of strength testing and other outcome measures for evaluating children with SMA. It will provide a framework for developing mechanisms whereby therapies may be tested efficiently and economically in SMA children. The prevalence and severe nature of this disease are compelling motives for attempting such work.

Back to Article Outline

7. Animal models of SMA 

A new exciting area of research is the development of animal models of the disease, in particular, mouse models of SMA. In contrast to humans, mice harbour only one copy of the Smn gene, which is equivalent to human SMN1. Most of the SMA mice that have been recently generated present with symptoms similar to those seen in SMA patients. They promise to further the understanding of the molecular basis of this disease [28]. The first mice model was generated by the group of Burghes [29] with disease symptoms and neuropathology similar to those of severe type I SMA in humans. They created transgenic mice that express human SMN2 and mated these onto the null Smn−/− background. The Smn−/−;SMN2 mice carrying one or two copies of the transgene have normal numbers of motor neurons at birth, but vastly reduced numbers by postnatal day 5, and subsequently die. On the contrary, the mice expressing eight copies of the SMN2 gene on the Smn−/− knockout background showed no overt SMA phenotype or muscle weakness, indicating that phenotypic severity can be modulated by SMN2 copy number. The other relevant outcome of this model is that motor neuron loss is a late onset phenomenon and occurs after birth in SMA mice.

A similar mouse model was generated by a Taiwanese group [30]. However, their transgenic mice harbouring SMN2 in the Smn−/− background knocked out of exon 7 showed the three phenotypes characteristically seen in SMA patients. The severity of pathological changes in these mice correlated with the amount of SMN protein that contained the region encoded by exon 7. A French mouse model of SMA [31] was generated via the deletion of the murine survival of motor neuron gene exon 7 directed to neurons, but not to skeletal muscle. This model provided evidence that motor neurons are the primary target of the gene defect. Moreover, the mutated SMN protein (SMNDeltaC15) was dramatically reduced in the motor neuron nuclei and caused a lack of gems associated with large aggregates of coilin, a coiled-body-specific protein. These results identified the lack of the nuclear targeting of SMN as the biochemical defect in SMA [31]. Another group [32] showed that the heterozygous mice lacking one Smn allele (Smn+/− mice) have Smn protein levels that are reduced by ∼50% in the spinal cord and show a marked loss of the cytoplasmic Smn pool and motor neuron degeneration resembling spinal muscular atrophy type III. These Smn heterozygous mice thus represent a model for the human disease.

Judith Melki (France) reported on a new animal model in which the deletion of murine SMN exon 7, the most frequent mutation found in SMA, has been restricted to either neurons or skeletal muscle by using the Cre-loxP recombination system [33]. In the ‘neuronal’ mutant, histological changes were consistent with muscle denervation pattern associated with marked changes of motor neurons and axons. In contrast, the ‘muscular’ mutant displays a muscular dystrophic phenotype characterized by the presence of necrotic fibres and regenerating myocytes without the muscle denervation process. The muscular mutant mice display ongoing muscle necrosis with a dystrophic phenotype leading to muscle paralysis and death at 1 month. The pathology in the muscle starts at 3 weeks, and at 4 weeks a dystrophic pattern with rapid progression is evident. The dystrophic phenotype is associated with elevated levels of creatine kinase activity, Evans blue dye uptake into muscle fibres, reduced amount of dystrophin and up regulation of utrophin expression, suggesting a destabilization of the sarcolemma components. However, sarcoglycan and dystroglycan are normally expressed. No defect of the basal lamina has been noted. There were no changes in the motor neurons or in the neuromuscular junction. These results suggest that both motor neurons and skeletal muscle are most likely involved in SMA pathogenesis, and it can be hypothesized that a similar mechanism is involved in motor neuron or skeletal muscle degeneration. Further molecular and morphological characterization of mouse models of SMA should have important implications for the development of therapeutic strategies in SMA.

Back to Article Outline

8. Review of neuroactive drugs 

Leonard van der Berg (The Netherlands) reviewed the potential drug treatments for amyotrophic lateral sclerosis (ALS) based on the pathophysiology of the disease. The precise pathogenesis of ALS is not understood, nonetheless, at least three principal mechanisms are advocated: excitotoxicity, oxidative damage, and lack of neurotrophic factors. Excitotoxicity is a process in which neuronal damage may occur if an excess of glutamate causes over-stimulation of glutamate receptors. Numerous clinical trials have tested the efficacy of anti-glutamergic substances. A multicenter trial has shown a three-month increase in survival of ALS patients treated with riluzole versus placebo group during an observation period of two years. Gabapentin has been tested in double blind placebo controlled trials and has shown positive effects in small phase II trials, but no significant effects were seen in two large phase III trials. Oxidative stress is considered a possible mechanism in ALS, as mutations in the gene for SOD1 underline 10% of familial ALS cases. SOD1 detoxifies superoxide radicals into hydrogen peroxide, which is then converted to water by glutatione peroxidase. Drugs that have anti-oxidative actions have been considered for trials in ALS. Vitamin E has not been proven effective in ALS patients and only marginally effective in transgenic SOD1 animals. CoQ10 is a substance that acts in the mitochondrial respiratory chain. It is currently being checked for effectiveness in ALS in a US randomised trial. Creatine has been shown to be effective in a dose dependent way in SOD1 mutant mice.

Giuseppe Vita (Italy) reviewed other neuroactive and neurotrophic factors potentially useful in SMA trials. The use of a glutamate inhibitor such as gabapentin in ALS and SMA trials comes from the assumption that glutamate-induced neurotoxicity is a common pathogenetic event for both acute and chronic neuronal death. It begins with increased Ca2+ entry into the neuron, but calbindin, parvalbumin, and mitochondria normally buffer excessive ion entry [34]. Uncontrolled Ca2+ entry induces several enzymes, into producing nitric oxide and superoxide, and leads to more free radical production, injuring DNA and cell membrane structures [35]. Although direct evidence for glutamate-induced cellular damage in SMA is lacking, some data, such as increased CSF glutamate levels in ALS and hypersensitivity of target-deprived motoneurons in SMA to normal concentrations of glutamate suggest that such indeed may be the case [10]. There is also evidence that, in a transgenic mouse model of ALS, SOD1 mutants mediate toxicity through an excitotoxic mechanism [36]. However, gabapentin cannot be considered a true glutamate inhibitor because it acts as an antagonist of Ca-voltage channels and increases the GABAergic neurotransmission, reducing the release of especially non-synaptic, GABA and glutamate. On the contrary, topiramate is a real glutamate inhibitor. It acts as an antagonist of AMPA-kainate receptors (more abundant in the spinal cord), Ca-voltage channels, and voltage-activated Na channels [37]. Topiramate was approved by the FDA as a treatment for epilepsy in December 1996 and it has since been used in adults and children without report of significant adverse reactions. At the moment, a 21-site trial of topiramate is underway in the US to evaluate its safety and effectiveness in the treatment of ALS patients. Treated patients receive 400 mg/day. Enrolment began in July 1999, and all sites are still enrolling patients. Nearly 300 people will be enrolled in this 1-year study.

Creatine is synthesized in the liver, kidneys, and pancreas from the precursors arginine, glycine, and methionine. It is also ingested by meat and fish. Phosphocreatine is synthesized by creatine kinase from creatine by adding a ‘high-energy phosphate’ which, upon hydrolysis, transfers its chemical energy to ATP, the universal energy currency in the cell. Creatine supplementation in humans leads to an increase of intracellular creatine and phosphocreatine, improving anaerobic performance of muscle, shortening muscle relaxation time, increasing fat-free body mass, and the cross-sectional area of all muscle fibre types. Recent evidences of neuron protection have been reported. Malonate and 3-nitropropionic acid (3-NP) are inhibitors of succinate dehydrogenase that produce energy depletion and lesions that closely resemble those of Huntington's disease in rats. Creatine produces significant protection against malonate and 3-NP neurotoxicity, as demonstrated by histochemical, biochemical and 1H magnetic resonance spectroscopy [38]. Systemic administration of MPTP produces parkinsonism by a mechanism involving impaired energy production. Oral supplementation of creatine produces significant protection against MPTP-induced dopamine depletion in mice, as demonstrated by immunostaining evidence of neurons in the substantia nigra [39]. Finally, creatine produces a dose-dependent improvement in motor performance and extends survival in SOD1 transgenic mice, and it protects mice from loss of both motor neurons and substantia nigra neurons at 120 days of age [40]. Creatine also prevents an increase in biochemical indices of oxidative damage. A double blind randomised placebo-controlled trial of the safety and efficacy of creatine in ALS is currently recruiting patients in US. One hundred and fourteen eligible subjects will be randomised to receive treatment for 6 months or placebo. The primary outcome measure is the change in upper extremity motor function. Secondary outcome measures include grip strength, motor unit number estimates, the ALS functional rating score-revised, and the rate of change of a well established biochemical marker of oxidative damage to DNA such as 8OH2′dG levels in urine. If creatine slows disease progression in ALS and is well tolerated, a phase 3 study, with survival as the primary outcome measure, will be initiated.

Xaliproden is an oral, non-peptide neurotrophic and neuroprotective agent, which is produced by Sanofi-Synthelabo, France. The preliminary results of a clinical trial in ALS were presented at the 11th International Symposium on ALS/MND on January 2001. While there was no definite change in survival, there was significant improvement in respiratory function and improvement in swallowing, arm strength, and daily functioning when compared to placebo. The medication was very well tolerated; side effects such as nausea, insomnia, diarrhea, and vertigo usually resolved after the first month. The manufacturer has not yet decided whether to ask the FDA for approval, or to pursue additional therapeutic trials first.

On January 25, 2001, Amgen-Regeneron Partners discontinued all clinical development of brain-derived neurotrophic factor (BDNF) for the potential treatment of ALS following notification that BDNF did not provide a therapeutic advantage to ALS patients in clinical trials. The Amgen–Regeneron subcutaneous study of BDNF did not show any evidence of efficacy as measured by the primary endpoint of reduction in mortality or the need for continuous mechanical ventilation. In the intrathecal trial, BDNF showed no statistical difference from placebo in any measure, including survival and ventilator use.

Back to Article Outline

9. Specific European clinical trial criteria 

Stephen Trueman (Italy) presented an overview of the principal criteria for EU-funded clinical trials and the issues raised by the participants. The specific criteria for clinical trials can be summarised as follows: (i) Project must be large scale. (ii) The trial outcome, in compliance with accepted ethical standards, must have potential benefit and be accessible to the patient. This reflects one of the more general requirements of the Vth framework criteria: the social impact of the project. That is, how the project actually changes the life of the patients. (iii) The project should make a substantial contribution to public health. (iv) The major focus of the trial should not be commercial profit. Clinical trials that ought to be exclusively funded by industry will not be considered for EC funding. This is a new criterion for this call, which on one hand seeks to finance those trials that would not be given backing by industry, and on the other hand supports research with a lower commercial interest, such as rare diseases. (v) The trial must be a phase III and IV clinical trial. Phases II and I will not be supported. (vi) Ethical criteria must be satisfied in particular: research must be justified in terms of potential benefits in relation to possible risk; all consent must be informed, authorisation or approval of local ethics committees, and national bodies must be obtained, and specification and reference of all relevant national and international regulations must be carried out.

In addition, the specific issues raised for consideration in a European SMART were as follows: (i) In the realm of rare diseases it was felt that large scale trials should have wide geographical coverage to be representative of Europe. (ii) Patient perception of the work must be nurtured. (iii) European shared-information is a key objective. (iv) The dissemination of the results should be pro-active. As with Myo-cluster [41], a user group should be formed early on. (v) Particularly strong management is required to avoid the inherent risk of failure due to any one partner not delivering results.

At the end of this session, all participants expressed a strong interest in participating in such a project proposal, and consideration of the criteria presented strongly reinforced the overall belief that the project might be financed.

Back to Article Outline

10. List of workshop participants 

Leonard van der Berg, Utrecht (The Netherlands)

Victor Dubowitz, (ENMC)

Brigitte Estournet-Mathiaud, Garches (France)

Susan Iannaccone, Dallas (USA)

Judith Melki, Evry (France)

Luciano Merlini, Bologna (Italy) organiser

Francesco Muntoni, London (UK)

Massimo Musicco, Milan (Italy)

Karen Rabb, Dallas (USA)

Sabine Rudnik-Schöneborn, Aachen (Germany)

Giovanni Savettieri, Palermo (Italy)

Haluk Topaloǧlu, Ankara (Turkey)

Stephen Trueman, Bologna (Italy)

Andoni Urtizberea, (ENMC)

Giuseppe Vita, Messina (Italy)

Thomas Voit, Essen (Germany)

Back to Article Outline

Acknowledgements 

The workshop was made possible thanks to the financial support of the European Neuromuscular Centre (ENMC) and ENMC main sponsors: – Association Francaise contre les Myopathies (France); Deutsche Gesellschaft fur Muskelkranke (Germany); Telethon Foundation (Italy); Muscular Dystrophy Campaign (UK); Muskelsvindfonden (Denmark); Princes Beatrix Fonds (The Netherlands); Schweizerische Stiftung fur die Erforschung der Musckelkrankheiten (Switzerland); Verein zur Erforschung von Muskelkrankheiten bei Kindern (Austria); Vereniging Spierziekten Nederland (The Netherlands) and ENMC associate member, Muscular Dystrophy Association of Finland. We are grateful to Professor Victor Dubowitz for his scientific help, and to Mr Michael Rutgers for the organizational assistance of the ENMC.

Back to Article Outline

References 

  1. Merlini L, Stagni SB, Marri E, Granata C. Epidemiology of neuromuscular disorders in the under-20 population in Bologna Province, Italy. Neuromuscul Disord. 1992;2:197–200
  2. Munsat TL, Davies KE. International SMA consortium meeting. (26–28 June 1992, Bonn, Germany). Neuromuscul Disord. 1992;2(5-6):423–428
  3. Lefebvre S, Burglen L, Reboullet S, et al.  Identification and characterization of a spinal muscular atrophy – determining gene. Cell. 1995;80(1):155–165
  4. Crawford TO, Pardo CA. The neurobiology of childhood spinal muscular atrophy. Neurobiol Dis. 1996;3(2):97–110
  5. Coovert DD, Le TT, McAndrew PE, et al.  The survival motor neuron protein in spinal muscular atrophy. Hum Mol Genet. 1997;6(8):1205–1214
  6. 38th ENMC International Workshop . Spinal muscular atrophy trial group 10-12 December 1995, Naarden, The Netherlands. Neuromuscul Disord. 1996;6(4):293–294
  7. Frey JR, Kaeser HE, Krause R. Therapeutic trials with an anabolic steroid and hexahydrocoenzyme Q4 in a case of spinal muscular atrophy. Eur Neurol. 1970;3(5):308–318
  8. Angelini C, Micaglio GF, Trevisan C. Guanidine hydrochloride in infantile and juvenile spinal muscular atrophy. A double blind controlled study. Acta Neurol (Napoli). 1980;2(6):460–465
  9. Tzeng AC, Cheng J, Fryczynski H, et al.  A study of thyrotropin-releasing hormone for the treatment of spinal muscular atrophy: a preliminary report. Am J Phys Med Rehabil. 2000;79(5):435–440
  10. Greensmith L, Vrbova G. Possible strategies for treatment of SMA patients: a neurobiologist's view. Neuromuscul Disord. 1995;5(5):359–369
  11. Gurney ME, Cutting FB, Zhai P, et al.  Benefit of vitamin E, riluzole, and gabapentin in a transgenic model of familial amyotrophic lateral sclerosis. Ann Neurol. 1996;39(2):147–157
  12. Miller RG, Gelinas DF, Moore DH. A phase III placebo-controlled trial of gabapentin in spinal muscular atrophy. Ann Neurol. 2000;48:468
  13. Tein I, Sloane AE, Donner EJ, Lehotay DC, Millington DS, Kelley RI. Fatty acid oxidation abnormalities in childhood-onset spinal muscular atrophy: primary or secondary defect(s)?. Pediatr Neurol. 1995;12(1):21–30
  14. Melki J, Sheth P, Abdelhak S, et al.  Mapping of acute (type I) spinal muscular atrophy to chromosome 5q12–q14. The French Spinal Muscular Atrophy Investigators. Lancet. 1990;336(8710):271–273
  15. Harpey JP, Charpentier C, Paturneau-Jouas M, Renault F, Romero N, Fardeau M. Secondary metabolic defects in spinal muscular atrophy type II. Lancet. 1990;336(8715):629–630
  16. Iannaccone ST, Browne RH, Samaha FJ, Buncher CR. Prospective study of spinal muscular atrophy before age 6 years. DCN/SMA Group. Pediatr Neurol. 1993;9(3):187–193
  17. Russman BS, Buncher CR, White M, Samaha FJ, Iannaccone ST. Function changes in spinal muscular atrophy II and III. The DCN/SMA Group. Neurology. 1996;47(4):973–976
  18. Deymeer F, Serdaroglu P, Poda M, Gulsen-Parman Y, Ozcelik T, Ozdemir C. Segmental distribution of muscle weakness in SMA III: implications for deterioration in muscle strength with time. Neuromuscul Disord. 1997;7(8):521–528
  19. Hausmanowa-Petrusewicz I. Electrophysiological findings in childhood spinal muscular atrophies. Rev Neurol. 1988;144(11):716–720
  20. van der Ploeg RJ, Fidler V, Oosterhuis HJ. Hand-held myometry: reference values. J Neurol Neurosurg Psychiatry. 1991;54(3):244–247
  21. Forrest M, Andersen B. Ordinal scale and statistics in medical research. Br Med J (Clin Res Ed). 1986;292(6519):537–538
  22. Brooke MH, Griggs RC, Mendell JR, Fenichel GM, Shumate JB, Pellegrino RJ. Clinical trial in Duchenne dystrophy. I. The design of the protocol. Muscle Nerve. 1981;4(3):186–197
  23. Cook JD, Glass DS. Strength evaluation in neuromuscular disease. Neurologic Clinics. 1987;5(1):101–123
  24. Beasley WC. Quantitative muscle testing: principle and applications to research and clinical services. Arch Phys Med Rehabil. 1961;42:398–425
  25. Noreau L, Vachon J. Comparison of three methods to assess muscular strength in individuals with spinal cord injury. Spinal Cord. 1998;36(10):716–723
  26. Schwartz S, Cohen ME, Herbison GJ, Shah A. Relationship between two measures of upper extremity strength: manual muscle test compared to hand-held myometry. Arch Phys Med Rehabil. 1992;73(11):1063–1068
  27. Iannaccone ST, Russman BS, Browne RH, Buncher CR, White M, Samaha FJ. Prospective analysis of strength in spinal muscular atrophy. DCN/Spinal Muscular Atrophy Group. J Child Neurol. 2000;15(2):97–101
  28. Monani UR, Coovert DD, Burghes AH. Animal models of spinal muscular atrophy. Hum Mol Genet. 2000;9(16):2451–2457
  29. Monani UR, Sendtner M, Coovert DD, et al.  The human centromeric survival motor neuron gene (SMN2) rescues embryonic lethality in Smn(-/-) mice and results in a mouse with spinal muscular atrophy. Hum Mol Genet. 2000;9(3):333–339
  30. Hsieh-Li HM, Chang JG, Jong YJ, et al.  A mouse model for spinal muscular atrophy. Nat Genet. 2000;24(1):66–70
  31. Frugier T, Tiziano FD, Cifuentes-Diaz C, et al.  Nuclear targeting defect of SMN lacking the C-terminus in a mouse model of spinal muscular atrophy. Hum Mol Genet. 2000;9(5):849–858
  32. Jablonka S, Schrank B, Kralewski M, Rossoll W, Sendtner M. Reduced survival motor neuron (Smn) gene dose in mice leads to motor neuron degeneration: an animal model for spinal muscular atrophy type III. Hum Mol Genet. 2000;9(3):341–346
  33. Cifuentes-Diaz CFT, Tiziano FD, Lacene E, Roblot N, Joshi V, Moreau MH, et al. Deletion of murine smn exon 7 directed to skeletal muscle leads to severe muscular dystrophy. J Cell Biol. 2001;152(5):1107–1114
  34. Gutmann L, Mitsumoto H. Advances in ALS. Neurology. 1996;47(4 Suppl 2):S17–S18
  35. Rothstein JD. Excitotoxicity hypothesis. Neurology. 1996;47(4 Suppl 2):S19–S25 discussion S26
  36. Trotti D, Rolfs A, Danbolt NC, Brown RH, Hediger MA. SOD1 mutants linked to amyotrophic lateral sclerosis selectively inactivate a glial glutamate transporter. Nat Neurosci. 1999;2(5):427–433
  37. Shank RP, Gardocki JF, Streeter AJ, Maryanoff BE. An overview of the preclinical aspects of topiramate: pharmacology, pharmacokinetics, and mechanism of action. Epilepsia. 2000;41(Suppl 1):S3–S9
  38. Matthews RT, Yang L, Jenkins BG, et al.  Neuroprotective effects of creatine and cyclocreatine in animal models of Huntington's disease. J Neurosci. 1998;18(1):156–163
  39. Matthews RT, Ferrante RJ, Klivenyi P, et al.  Creatine and cyclocreatine attenuate MPTP neurotoxicity. Exp Neurol. 1999;157(1):142–149
  40. Klivenyi P, Ferrante RJ, Matthews RT, et al.  Neuroprotective effects of creatine in a transgenic animal model of amyotrophic lateral sclerosis. Nat Med. 1999;5(3):347–350
  41. Merlini L. Myo-Cluster. Neuromuscul Disord. 2001;11:101–102

PII: S0960-8966(01)00272-3

Neuromuscular Disorders
Volume 12, Issue 2 , Pages 201-210, February 2002

Article Tools