Amyotrophic Lateral Sclerosis. 2005; 6: 202–212
REVIEW ARTICLE
Clinical trials in ALS: A review of the role of clinical and neurophysiological measurements
˜ O COSTA1 & MICHAEL SWASH4 MAMEDE DE CARVALHO1,2, JOA 1
Department of Neurology, Hospital de Santa Maria, Lisbon, 2Laboratory of Electromyography, Centro de Estudos Egas Moniz, Faculty of Medicine, Institute of Molecular Medicine, Lisbon, Portugal, 3Institute of Neuroscience, Barts, and 4The London Department of Neurology, Royal London Hospital, Queen Mary University of London, London, UK
Abstract We have reviewed all the published clinical trials of ALS and, from those considered sufficiently large, and containing a control group, we have evaluated their methodology with regard to statistical power. This implies a critical analysis of the endpoint measurements. We have concluded that clinical endpoints used in clinical trials of ALS have frequently been insufficiently sensitive, non-linear, or even not intuitively highly relevant to the disease. We suggest that the ALS-FRS, perhaps also MUNE and the Neurophysiological Index, may be the best measures currently available. These techniques have complementary characteristics that allow them to be used to address different aspects of the disease and its treatment in various trials designs. In the past some trials may have failed to demonstrate a treatment effect because the chosen endpoint measures and the trial design were inappropriate.
Key words: Motor neurone disease, amyotrophic lateral sclerosis, clinical trials, neurophysiology, clinical measurement
Introduction The devastating nature of amyotrophic lateral sclerosis (ALS) has led to a major effort to find a treatment capable of arresting the disease. However, although more than 100 trials have been reported (1), only one medication, riluzole, has shown a positive effect, prolonging survival for a few months. The design and implementation of clinical trials in ALS raises many scientific and ethical problems (1,2). Recruitment of suitable patients is a major issue. The number required depends on several aspects of trial design, especially the sensitivity of the measurement in detecting change, its standard variation in the studied population, the number of dropouts and the duration of the trial (3). Another problem concerns the likely importance of early diagnosis (4) so that treatment with the test drug can be commenced before a large number of motor neurons are irreversibly lost. A randomized, parallel, placebo-controlled, double-blind study is considered the gold standard in trial design. However, a large number of open trials, some with crossover design, have been carried out in the past. Many different drugs have shown activity in the SOD1 (G93A) mouse model. However, despite these
encouraging results, transposing these concepts to human ALS has proven generally disappointing. Another approach to finding potentially useful drugs for testing in human ALS in definitive trials would be to study a considerable number of compounds in small exploratory clinical trials with ALS patients, before developing a large Phase III study design. For this last strategy, it would be essential to choose very sensitive measurements, which could give information about effectiveness in a small group of patients. We set out to review all clinical trials in ALS conducted to date. Those fulfilling predefined criteria for quality were selected for evaluation of progression as defined by the measurements used in the trials. A ‘metanalysis-like approach’ was applied to obtain the mean values of progression for the different measurements regarding their potential future utility in exploratory clinical trials. Methods Search strategy for identification of studies Our literature review used the following search strategy: relevant trials were identified in the Cochrane Central Register of Controlled Trials,
Correspondence: M. Swash, Department of Neurology, The Royal London Hospital, London E1 1BB, UK. E-mail:
[email protected] (Received 7 March 2005; Accepted 5 July 2005) ISSN 1466-0822 print/ISSN 1471-180X online # 2005 Taylor & Francis DOI: 10.1080/14660820510011997
Clinical trials in ALS Cochrane Database of Systematic reviews, Database of Abstracts of reviews of Effectiveness (in Cochrane Library issue 4, 2003), and MEDLINE (1966/12/ 2004). Titles, keywords and abstracts of the citations downloaded from the electronic searches were screened, and full copies of reports of all potentially acceptable trials were obtained for study. The reference lists of published clinical trials and ALS review articles were also searched. The database search strategy is detailed below: 1. 2. 3. 4. 5. 6. 7. 8.
MEDLINE and Cochrane Library: -exp/Motor Neuron Disease Amyotrop* AND Later* AND Scleros*.tw 1 OR 2 random* OR placeb* OR blind*.tw exp/Randomized Controlled Trial 4 OR 5 3 AND 6 limit 7 to human
Criteria for selecting high-quality trials for extracting pertinent information regarding progression We included all randomized trials of patients with ALS, regardless of the clinical characteristics of the patients, the type of interventions evaluated and outcomes measured. Abstracts, letters, and full papers written in languages other than English and French were not considered. Since ALS is a chronic progressive disease we excluded trials with a crossover design because of the possibility of carry-over effects. We also excluded trials with treatment durations less than six months since there is a consensus (2) that such short trial periods are not likely to yield meaningful data. Because of the possibility of performance bias, trials not using a double-blind design were excluded. Furthermore, we excluded trials utilizing fewer than 30 patients per treatment arm, since we considered these were probably too small to give meaningful information. We evaluated the baseline data and changes in the measurement of progression in the placebo arm of the selected clinical trials in the original publications and, when available, in the Cochrane review. Since few studies utilized neurophysiological measurements we accepted studies utilizing these methods that included less than 30 patients. In some instances neurophysiological measurements were published subsequent to the main clinical data. In studies in which no therapeutic effect was observed, merged data from treatment and placebo arms were sometimes analysed, when these were available. We also included several prospective studies of neurophysiological measures in ALS, performed outside the context of a clinical trial.
203
order to identify potential trials according to the specified criteria. Disagreements were resolved by discussion.
Results Using the search criteria described we identified 56 full reports of randomized clinical trials in ALS. We analysed 24 trials, and excluded 32 according to our prior entry criteria (Table I). Information regarding the clinical and neurophysiological endpoint measurements used in these trials is summarized below. Cyclosporine – 1988 (5) This was a small study (38 in the placebo group and 36 in the experimental group) with a negative result (considering all patients as a single group). The primary outcome measure was the Appel scale. The sample size was calculated based on the relative risk of progression on the Appel scale of 22 points over 48 weeks (a50.05). The placebo group showed a non-significant higher rate of progression (3.9+/ 20.5/month) than the control group (3.4%+/ 20.3%). BCAA – 1993 (6,7) A total of 121 patients was involved (60 in the placebo group). The active and the control groups were not well matched. Sample size was calculated based on a reduction of 20% in the proportion of patients becoming non-self supporting at 1 year (power 90%; a50.05). The primary outcome included death and changes in three disability scores (MRC scale, Norris scale and Appel scale). The duration of treatment was 12 months. There was an excess death rate among the subjects given BCAA, and this drug had no effect on the disability scores. Growth hormone – 1993 (8) This was a small study (75 patients). The active and the control groups were not well matched and there were a number of dropouts. The primary endpoints included the Tufts Quantitative Neuromuscular Examination (TQNE) and the MRC strength score. The number of patients included was based on an expected change of 1 SD in any defined variable (power 95%; a50.05). The treatment was given for 18 months. No effect was found on survival or strength. Immunosuppression – 1994 (9)
Methods of literature review Two reviewers (JC, MC) independently assessed the studies identified by the search strategy in
Only 61 patients were included (30 in the active group). A large number of outcomes was considered (death, and changes in quantitative dynamometry,
204
M. de Carvalho et al.
Table I. Summary of excluded randomized clinical trials. These studies were defective for the reasons shown in the column headings. Reasons for exclusion Drug
Year
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
1971 1974 1978 1981 1984 1984 1984 1985 1986 1986 1986 1986 1988 1989 1989 1989 1991 1992 1992 1992 1992 1993 1993 1993 1994 1994 1996 1996 1997 1998 1998 2003
Isoprinosine (51) Guanidine (41) Tilorone (52) Amantadine and Guanidine (53) Gangliosides (54,55) Gangliosides (56,57) TRH (58) Levamisol (59) TRH (60) TRH (61) TRH (62) Octaconasol (63) BCAA (7,64) Gangliosides (65) L-threonine (7,66) BCAA (7,67) L-threonine (7,68) L-threonine (7,69) BCAA (7,70) Threonine (7,71) Threonine (7,72) Dextromethorphane (73) ORG2766 (74) Lamotrigine (75) Selegiline (13,76) Selegiline (13,77) Nimodipine (78) Dextromethorphan (79) Dextromethorphan (80) Methylcobolamin (81) Reduced glutathione (13,82) Lamotrigine (84)
Open trial
7
7
7 7
Few patients
Short duration
7 7 7 7 7 7 7 7 7 7 7 7 7 7
7
7 7 7 7 7 7 7 7 7
Cross-over
Other
7
7 7 7 7 7
7 7 7 7 7
7 7
7
7
7
7
7
7 7
7
7 7
7 7
7 7
7 7
7 7 7
7
7 7 7 7 7
*
Results presented in an abstract or in trials aimed to relieve symptoms were excluded. *Randomization uncertain; the ‘control group’ received a low dose of the trial drug.
manual muscle testing, functional tests and an activity index). Somewhat surprisingly, power was estimated as detecting a change in outcome in as few as 25% of the treated patients, compared with the control subjects (power 80%; a50.05). Patients were followed for two years, or until death or respirator dependency. Survival and motor function were not different in the two groups. Riluzole – 1994 (10,11) In this study 155 patients were included (78 in the placebo arm). Survival (equal to death or tracheostomy and assisted ventilation) and change in the functional status scale (modified Norris scale) were the primary endpoints. This study was designed to observe a change in the expected survival from 55 to 85% at 1 year (power 90%; a50.05). Survival was increased by 38.6% in the treated group and the Norris scale showed a non-significant slower progression (33.4%). This study led to a larger trial (number 11 below), and a further trial in older people with ALS, and in those with advanced disease (number 19 below) (83).
Acetylcysteine – 1995 (12,13) One hundred and eleven patients were recruited (56 in the placebo group). The primary outcome was survival (equal to death or tracheostomy and assisted ventilation). The study was designed to detect a change of 50% in survival at 1 year (power 80%; a50.05) but no difference was detected in the data after adjustment for baseline prognostic factors, although the treated group had an overall mortality 26% lower than the placebo group. Subcutaneous rHCNTF – 1996 (14,15) This was a large study. It included 730 patients, who were followed for nine months. There were two active treatment groups with different drug regimens, and one placebo group containing 245 patients. The number of dropouts was significant, in particular for the high-dose group (26.1%). The expected differences, as well as the power the study and the alpha value were not specified. The primary outcomes were the rate of muscle strength loss as determined by maximal voluntary isometric contraction (MVIC)-combined megascore. In the
Clinical trials in ALS placebo group the rate of change was 0.17+/20.05/ month, but there was no difference between the placebo and treated groups (p50.67). Subcutaneous rHCNTF – 1996 (15,16) This 6-month trial contained four treatment arms and included 570 patients (123 in the placebo group). All subjects were observed in a 2–7 month lead-in period before treatment commenced. The primary endpoint was the change in a megascore calculated from MVIC of 18 muscles and the forced vital capacity. The number of patients entered was set to detect a difference of 35% (a50.05, power 80%). The groups were well matched and the number of dropouts was small. At all doses tested, the drug had no beneficial effect.
205
rhIGF-I – 1997 (20,21) This study lasted for nine months and included 266 patients. There were two active treatment groups with different drug dosages, and one placebo group containing 90 subjects. Only 141 patients finished the study. The primary outcome measure was the rate of change of the Appel score. No details are available regarding the rate of change designed into this endpoint, which must have been based on the change expected in the placebo group, and the b and a for the protocol were not disclosed. No difference was found between the active treatment and the placebo groups, although the high-dose active group (0.1 mg/kg/day) progressed 26% more slowly than the placebo group. Selegiline – 1998 (13,22)
BCAA – 1996 (7,17) This study included 95 patients (31 received BCAA, 32 L-Threonine and 32 placebo). The primary outcome was a modified MRC score and a combined Z score of the MVIC of several muscles. The study was designed to detect an outcome difference of 50% at six months (a50.05, power 80%), but no treatment effect was noted. In the placebo group the MRC score declined 11.3% (11.6) in six months. The rate of loss in pulmonary function was higher in the active treatment groups. Gabapentin – 1996 (18) This study continued for six months and included 152 patients (79 in the active group). The groups were not well matched for age. The number of dropouts was rather large (25%). The primary outcome was the mean slope of the arm MVIC megascore. The study was set to detect a difference of 50% at six months (power 80%, a50.05). The mean rate of change was 24% lower in the active group, but this was not significant. Riluzole – 1996 (11,19) This large study continued for 18 months, and included 959 patients; 122 in the placebo group and the rest in three separate dosage groups. The primary outcome was tracheostomy-free survival. A minimum of 150 patients was entered in each treatment group. The trial was powered to detect a hazard ratio of 0.67 for the 100 mg/day riluzole group relative to the placebo group, with a one-side log-rank test at a50.05, a power 85%, and an assumed 35% survival in the placebo group at 18 months. Survival at 18 months was 50.4% in the placebo group and 56.8% in the treated group (100 mg/day). This difference was significant, indicating a therapeutic effect of riluzole.
This trial included 133 patients (66 in the placebo group) for six months, the groups were well balanced and the number of dropouts was acceptable. The primary outcome was the Appel ALS total score. Two analyses were used: variance analysis (intentto-treat model) and Kaplan-Meier survival curves, using an overall change of 22 points in the Appel ALS total score (16% of the total score, in six months, 2.7%/month or 3.7 points in that score/ month). These changes were defined on the basis that they represent a major change in a patient’s functional status (powerv80%, a50.05). A mean rate of change of 3.5 points/month (2.6%) in the placebo group and 3.4 in the treated group was observed, but there was no difference in the survival analysis (pw0.05). The number of patients included in the study was less than the number necessary (180) to reach the defined power of 80%, although the exact number was not given. rhIGF-I – 1998 (21,23) This was a nine-month trial that included 183 patients (124 in the active group, 0.1 mg/kg/day). Patients in the active group had a higher mortality risk than those in the placebo group. The primary outcome was the change in the Appel score, but no power calculation or alpha error was disclosed. The placebo score decreased by a mean of 25.2+/22.3 (18.8%+/21.7), (mean of 2.8/month, 2.1%/month). The active treatment group showed less change, 21.9+/21.5 (16.3%+/21.1), or 13.1% less, but this difference was not significant. BDNF – 1999 (24) Over nine months 1135 patients were studied in this trial (748 in the active treatment groups, consisting of two dose-regimens). The patients in the active treatment groups had longer disease duration. Some patients in each group were taking riluzole, but the
206
M. de Carvalho et al.
impact of this on the results was not analysed. The primary outcomes were survival at nine months (a50.025, power 74% to detect a difference of 10% in survival) and FVC decline (a50.013, power 90% to detect an absolute treatment difference of 5.7%). Over six months the FVC declined 13.8%+/21.0 in the placebo group. The decline in the active group (15.8%+/21.1) was higher. The nine months adjusted survival was 84.8% for the placebo group and 86.7% for the active treatment group. This was not a significant difference, although it was suggested that patients with early respiratory impairment might benefit. Recombinant interferon beta 1-a – 2000 (25) The study consisted of a six-month pre-treatment observation period followed by a second six-month period of active trial. It included 61 patients (30 in the placebo arm). The groups were well matched, except for gender. During the trial period the use of riluzole was not permitted. The number of dropouts was very large (25). The primary outcome was the number of patients becoming non-self supporting during this 12-month period. The study was designed to detect a 30% absolute difference between active and placebo groups (a50.05, power 80%). No difference was found. Gabapentin – 2001 (26) This study included 204 patients (102 in each arm) and lasted for nine months. The active group had a shorter duration of symptoms; otherwise the groups were well matched. The number of dropouts was considerable (27.9%). The primary outcome was the mean slope of the MVIC arm megascore. A 35% difference in MVIC slope in the two groups was considered significant (a50.05, power 80%). There was no difference in the change per week between the placebo group (0.021+/20.022) and that in the active treatment group. a2Tocopherol – 2001 (13,27) This 12-month trial included 289 patients. Apart from age, the active (a2tocopherol 500 mg b.i.d. with riluzole 50 mg b.i.d.) and control (placebo and riluzole 50 mg) groups were well matched. However, the number of dropouts was very large (50.1%). The primary outcome was a modified Norris limb scale (maximum 63 points). The number of patients included was calculated on the basis of a change of 7 points (11.1%) at the end of the trial (a50.05, power 80%). The placebo group scores decreased 14.4+/212.5 (27.8%+/223.8; 2.3%/month), and the treated group 12.5+/212.9 (23.7%+/224.5, 2.0/month), which represents a non-significant difference.
Riluzole – 2002 (83) In this study 168 patients were randomized either to riluzole or to placebo, and treated for 18 months. All subjects had ALS and were either at an advanced stage or were aged more than 75 years. The study was not powered to detect a difference in survival between the two groups. No significant side-effects were observed. Creatine – 2003 (28) This 16-month trial included 175 patients (88 treated). A sequential trial design was used with death, persistent assisted ventilation, or tracheostomy as primary endpoints. Concurrent medication with riluzole was part of the inclusion criteria, and 10 g/day of creatine monohydrate or placebo was given to the active and placebo group, respectively. Sample size calculation was based on an expected 25% increase in the active group (a50.05, power 90%). No difference was found. The possibility of additional self-medication with creatine, outside the trial design, could not be excluded. Topiramate – 2003 (29) This trial lasted for 12 months and included 296 patients, in a 2:1 ratio for active treatment and control groups. Patients could take riluzole if the dose had been stable for at least two months before the initial, baseline visit. The number of dropouts was very large (44.5%). The primary endpoint was the rate of change of the MVIC upper limb megascore. A 35% difference was used to estimate the number of patients included (a50.05; power 80%). The topiramate group showed a steeper slope of decline (33.3%) than the placebo group. Xaliproden – 2004 (30) This 18-month trial included two studies: study 1, without riluzole (active drug, 1 mg and 2 mg, versus placebo (286 patients), with a total of 867 patients; study 2 with riluzole (active drug, 1 mg and 2 mg, both with riluzole), versus placebo with riluzole (406 patients), with a total of 1210 patients. Stratification by the clinical site of onset within each treatment arm was peformed for each study. Two primary endpoints were defined; first, time to death, tracheostomy, or permanent assisted ventilation and, second, time to vital capacity decline below 50% of the predicted value, or permanent assisted ventilation (before and after adjustment for prespecified prognostic factors. For study 1, the number of patients was calculated to detect a 38% risk reduction in mortality, taking into account adjustments for two dose group comparisons with placebo and two primary
Clinical trials in ALS endpoints (a50.0125, power 85%). For study 2, the number of patients was calculated to detect a 31% risk reduction in mortality for those receiving xaliproden-riluzole combination (a50.0125, power 85%). Overall in the two studies the groups were well matched, treatment compliance was excellent and the number of dropouts was small. In study 1, the group receiving 2 mg of xaliproden, the time to vital capacityv50% (without permanent assisted ventilation) was significantly increased (pv0.045, adjusted analysis), but no difference was found regarding the other primary endpoints. In study 2, significant results were obtained for the 1 mg dose, concerning time to vital capacityv50% (without permanent assisted ventilation), (pv0.003, adjusted analysis), time to vital capacityv50% or permanent assisted ventilation (pv0.015, adjusted analysis), and time to vital capacityv50% or death (pv0.01, adjusted analysis). The clinical value of these differences seemed too small to justify further evaluation of this drug. In addition, there was evidence that a combination of riluzole and xaliproden slightly decreased survival. a2Tocopherol – 2004 (31) This 18-month trial included 160 patients (77 in the placebo arm). Apart from age the active (a2tocopherol 500 mg b.i.d. and riluzole) and control (placebo and riluzole) groups were well matched. Before trial onset, most patients were taking riluzole and 23% vitamin E. The primary outcome was survival (death, tracheostomy or permanent assisted ventilation). The study was powered to detect a 50% improvement in the treated group, but no more details are given regarding this issue. The number of dropouts was high. Overall the survival rate was statistically similar in the two groups.
Creatine – 2004 (32) This six month trial included 104 patients (50 treated), the patients allocated to active treatment received creatine 20 g/day for five days followed by 5 g/day; all subjects in the trial could take riluzole if the dose used had been stable for at least two months prior to the baseline visit (about half in each group). The number of dropouts was acceptable. The primary outcome measure was MVIC of eight upper extremity muscles. The sample size was calculated from the slope of the expected average change, in order to detect a 50% difference in the treated group (a50.05, power 80%). Treatment with creatine had no significant effect on the rate of change during the trial, estimated as less than 23%. We used the data from the placebo arm of these 24 clinical trials to evaluate the potential future utility of the endpoint measurements used. These measurements were the Norris, Appel and ALS-FRS functional scales, manual muscle testing (based on MRC score), MVIC, forced vital capacity (FVC)/ vital capacity (VC), compound muscle action potential (CMAP) amplitude, and motor unit number estimation (MUNE). For the neurophysiological measurements, data from papers published after the initial trial reports were also used; in some of these later reports the data from the placebo and non-effective treatment arms were merged, and we accepted these merged data. Our own prospectively collected published data were included (33,34) (Table V). This information is summarized in Tables II, III, IV and V. Since we were concerned with measurements in living patients we did not evaluate survival. After calculating the mean value and the coefficient of variation at baseline and calculating the standard deviation for each measurement, as well as the mean change at six months, we calculated the number of
Table II. Trials using functional scales as endpoints. Data at Baseline Trial Norris BCAA (6,7) Riluzole (Limb) (10,11) Riluzole (Bulbar) (10,11) Riluzole (Limb) (11,19) Riluzole (Bulbar) (11,19) BCAA (7,17) Vitamin E (Limb) (13,27) Vitamin E (Bulbar) (13,27) ALS-FRS BNDF (24) Creatine (28) Topiramate (20) Appel rhIGF-I (20,21) Selegiline (13,22) *Estimated.
207
Change (%)
Mean
SD
CV
6 months
56.6 40.8 30.1 44.1 30.3 75.2 52.4 33.1
6.1 16.0 11.0 15.2 9.6 9.0 10.3 7.6
0.11 0.39 0.37 0.35 0.32 0.12 0.20 0.23
17.1 22.9* 13.3* 18.2* 12.2* 10.1 14.0* 12.0*
30.1 30.7 30.3
5.2 4.9 5.9
0.17 0.16 0.20
13.0* 20.0* 18.2*
71.5 54.9
14.2 11.4
0.20 0.21
13.2 15.7
208
M. de Carvalho et al.
Table III. Trials using measurement of strength as an endpoint. Data at baseline Trial Manual muscle testing BCAA (6,7) Riluzole (10,11) Acetylcysteine (12,13) Riluzole (11,19) BCAA (7,17) Interf b 1-a (25) bVitamin E (13,27) Measurement with MVIC** BCAA (7,17) Yuen and Olney (85) Gabapentin (26)
Change (%)
Mean
SD
CV
6 months
85.0 79.1 85.0 79.1 184.0 129.0 130.8
11.0 19.0 11.0 16.8 23.0 14.5 16.0
0.13 0.24 0.13 0.21 0.13 0.11 0.12
16.5 17.0* 9.0* 14.0* 11.3 11.0 9.0*
16.3 24.2 79.4
11.1 5.5 53.7
0.68 0.23 0.68
36.2 20.0 21.0*
*Estimated; **Different methods.
patients needed in a parallel design trial to detect a 50% change in the rate of progression (power 80%, a50.05) considering each of the different measurements as potential primary endpoints. The following formula was applied to calculate the number needed to treat (3): 2s2 z1{a=2 zzp 2 D2 In this formula, standard deviation of the measurement at baseline and the expect rate of change as well as the hypothesized drug effect are considered. We did not include dropouts or estimators for the possible intersubject variabilities in our calculation. Table VI summarizes these results. The smallest number needed to treat is derived from trials using ALS-FRS and NI as endpoint measures.
Discussion Survival seems intuitively a simple event to capture in a clinical trial of ALS. However, there are potential problems even with so evident a measure. For example, it is survival rather than death that is the important outcome. As an endpoint, survival requires an adequate trial duration in order to allow
for enough deaths to occur in the control group to develop statistical power (35). Interventions such as non-invasive ventilation (36) and gastrostomy (37) probably influence survival, thus reducing the specificity of this endpoint. Measurement of muscle strength has often been considered an appropriate measure since it addresses the main problem in this condition, progressive weakness. This measurement, however, is also problematic since it is affected by other factors, especially technical matters, such as test-retest reproducibility, and also practicality, sensitivity, patient cooperation, and motivation (38). Respiratory tests assess the function of a particular set of muscles that are crucial for survival in ALS, but FVC does not necessarily have relevance to other muscle groups. Forced vital capacity has limitations in patients with bulbar problems (39). Sniff nasal pressure (40) can be useful, but questions remain concerning its sensitivity. Nocturnal oximetry can be affected by other coincidental conditions, such as sleep apnoea (39). A number of functional scales have been developed in ALS to reflect the overall clinical situation. The Norris (41) and Appel (42) scales include both subjective and objective features, but there is a large experience in applying them in clinical trials.
Table IV. Measurement of forced vital capacity – vital capacity as% predicted value. Data at baseline
Change (%)
Trial
Mean
SD
CV
6 months
BCAA (6,7) Acetylcysteine (12,13) BCAA (7,17) CNTF (14,15) CNTF (15,16) BNDF (24) Interferon b 1-a (25) Vitamin E (13,27) Creatine (28) Topiramate (29)
80.0 80.0 72.7 74.4 80.9 87.5 94.6 91.8 89.0 86.0
24.0 24.0 21.4 29.7 18.9 19.1 13.1 22.1 18.4 24.5
0.30 0.30 0.29 0.40 0.21 0.22 0.14 0.24 0.21 0.28
26.3 13.0* 11.3 13.2* 16.4 15.8 20.0* 10.0* 14.0* 17.0*
*Estimated.
Clinical trials in ALS
209
Table V. Neurophysiological measurements as endpoints. Data at baseline Trial M-wave amplitude Yuen and Olney (86) Felice (86) Bromberg et al. (7,17,87) – site 1 Bromberg et al. (7,17,87) – site 2 De Carvalho and Swash ** Motor Unit Number Estimation Yuen and Olney (85) Felice (82) Bromberg et al. (7,17,87) – site 1 Bromberg et al. (7,17,87) – site 2 Shefner et al. (29,88) De Carvalho and Swash(34) Neurophysiological Index De Carvalho and Swash (33) + De Carvalho and Swash (33) ++ De Carvalho and Swash (34)
Change (%)
Mean
SD
CV
6 months
6.43 5.7 7.8 11.1 9.4
2.66 2.6 3.6 3.6 3.8
0.41 0.46 0.46 0.32 0.40
21.6 24.6 16.7 27.0 16.7
0.47 0.88 1.05 0.94 0.44 0.73
48.6 44.0 29.7 38.3 23.2 28.2
0.53 0.28 0.57
22.0 50.3 26.9
94.4 56.5 239 269 58.7 55.0 2.66 2.88 2.71
44.1 49.8 251 252 25.7 40.3 1.42 0.82 1.54
**Unpublished; +: Patients with slow progression; ++: Patients with quick progression.
(46). No biological measure so far available seems likely to fulfil these demands in trials of ALS. The best such measure currently available seems to be the ALSFRS or ALSFRS-R (47), a functional, patient-related scale that includes scaled responses to questions related to speech, salivation, swallowing, handwriting, feeding, dressing and hygiene, turning in bed, walking, climbing stairs, and breathing that was derived arbitrarily from clinical intuition and then tested in clinical trials and in a number of prospective studies. Is there anything to be gained by including more specific data from clinical neurophysiology in the clinical measures that might be used in clinical trials in ALS? The NI is a neurophysiological measure made up empirically from the M wave amplitude, the distal motor latency and the F-wave frequency (48). It represents an index based on these simple, and readily available measurements, and it has been shown to have validity in that it is sensitive to change during the course of ALS (33,34). There is a correlation between changes in NI and changes in
ALS-FRS (43), described more recently, is a simple, sensitive scale that can be applied by telephone interview (44). Quality of life has been recognized as an important issue in ALS. The incorporation of quality of life scales in clinical trials has therefore been suggested (45). However, quality of life scales are recommended as additional items and not as primary endpoints of efficacy in clinical trials (46). We did not address the problem of drugs having a negative impact on the progression of ALS, as observed in the topiramate trial, since our target was to look for positive effects. Our assessment of the trials reviewed suggests that too many of them were insufficiently powered to obtain statistically meaningful results. The problem of dropouts was insufficiently addressed in calculating power in several of these studies. In others the measure chosen as an indicator of drug efficacy was not sensitive, or was non-linear through different points on the scale (e.g. the Appel scale). While no single measure is likely to address all possible criticisms, such a measure must be both sensitive and relevant Table VI. Calculation of sample size using different endpoints.
Functional scales Norris ALS-FRS Appel Manual muscle testing MVIC FVC-VC Neurophysiological measures M-wave amplitude MUNE NI
Mean
SD
CV
Change (%) 6 months
Sample size*
72.5 30.4 63.2 110.3 40.0 83.7
16.7 5.4 13.3 16.9 21.2 21.8
0.23 0.18 0.21 0.15 0.53 0.26
14.7 17.1 14.5 12.5 25.7 14.7
154 67 132 95 267 197
8.1 128.8 2.73
3.3 95.1 1.26
0.41 0.74 0.46
21.3 35.3 33.1
230 274 122
MUNE: Motor Units Number Estimation. NI: Neurophysiological Index. *Number of patients per arm to observe a 50% lesser change, in six months, for each of the tested measurements, with power 80% and alpha of 0.05.
210
M. de Carvalho et al.
the ALS FRS (34) but the NI represents an objective neurophysiological measure, calculated on data from studies of the ulnar innervation of the abductor digiti minimi muscle, that may complement functional changes in handicap as expressed by the ALS FRS, and which may be a useful index of change in Phase I or Phase II clinical trials (48). MUNE, a direct measurement of the number of functional motor units in a muscle, has high sensitivity but its utility might be diminished by the high inter-subject variation (49). In general, neurophysiological measures have been neglected in thinking concerning endpoints in clinical trials, but they are accurate and replicable, and they deserve more serious consideration (50). They are potentially relevant because they assess the motor system quantitatively, and thus address the issue of measurement of the portion of the nervous system that is most vulnerable in ALS. This review highlights a number of serious defects in trial design in previously published clinical trials in ALS. These should be avoided in future trials, and in their analysis. Much of the problem seems to concern the sensitivity of outcome measures, and more work on this issue is needed. Future trials need to be shorter, and to involve smaller numbers of patients. Acknowledgment This work was supported by grants from the ‘Fundac¸a˜ o de Cieˆ ncia e Tecnologia’ (POCTI/ CBO/43952/2002), Portugal.
11.
12.
13.
14.
15.
16.
17.
18.
19.
References 1. Turner MR, Parton MJ, Leigh PN. Clinical trials in ALS: an overview. Sem Neurol. 2001;21:167–75. 2. Miller RG, Munsat TL, Swash M, Brooks BR. Consensus guidelines for the design and implementation of clinical trials in ALS. World Federation of Neurology committee on research. J Neurol Sci. 1999;169:2–12. 3. Matthews JNS, ed. An Introduction to Randomized Clinical Trials. London: Arnold, 2000. 4. Swash M. Early diagnosis of ALS/MND. J Neurol Sci. 1998;160(Suppl):S33–6. 5. Appel SH, Stewart SS, Appel V, Harati Y, Mietlowski W, Weiss W, Belendiuk GW. A double-blind study of the effectiveness of cyclosporine in amyotrophic lateral sclerosis. Arch Neurol. 1988;45:381–6. 6. The Italian ALS Study Group, Branched-chain amino acids and amyotrophic lateral sclerosis: a treatment failure? Neurology 1993;43:2466–70. 7. Parton M, Mitsumoto H, Leigh PN. Amino acids for amyotrophic lateral sclerosis/motor neuron disease. In: The Cochrane Library, Issue 4, 2004. Chichester, UK: John Wiley & Sons, Ltd. 8. Smith RA, Melmed S, Sherman B, Frane J, Munsat TL, Festoff BW. Recombinant growth hormone treatment of amyotrophic lateral sclerosis. Muscle Nerve 1993;16: 624–33. 9. Drachman DB, Chaudhry V, Cornblath D, Kuncl RW, Pestronk A, Clawson L, et al. Ann Neurol. 1994;35:142–50. 10. Bensimon G, Lacomblez L, Meininger V, and the ALS/ Riluzole study group. A controlled trial of riluzole in
20.
21.
22.
23.
24.
25.
26.
amyotrophic lateral sclerosis. New England Journal of Medicine 1994;330:585–91. Miller RG, Mitchell JD, Moore DH. Riluzole for amyotrophic lateral sclerosis (ALS)/motor neuron disease (MND). (Cochrane Review). In: The Cochrane Library, Issue 4, 2004. Chichester, UK: John Wiley & Sons, Ltd. Louwerse ES, Weverling GJ, Bossuyt PM, Meyjes FE, de Jong JM. Randomized, double-blind, controlled trial of acetylcysteine in amyotrophic lateral sclerosis. Arch Neurol. 1995;52:559–64. Orrell RW, Lane JM, Ross MA. Antioxidant treatment for amyotrophic lateral sclerosis/motor neuron disease (Cochrane Review). In: The Cochrane Library, Issue 4, 2004. Chichester, UK: John Wiley & Sons, Ltd. ALS CNTF Treatment Study Group, A double-blind placebo-controlled clinical trial of subcutaneous recombinant human ciliary neurotrophic factor (rHCNTF) in amyotrophic lateral sclerosis. Neurology 1996;46:1244–9. Bongioanni P, Reali C, Sogos V. Ciliary neurotrophic factor (CNTF) for amyotrophic lateral sclerosis/motor neuron disease (Cochrane Review). In: The Cochrane Library, Issue 4, 2004. Chichester, UK: John Wiley & Sons, Ltd. Miller RG, Petajan JH, Bryan WW, Armon C, Barohn RJ, Goodpasture JC, et al., and the rhCNTF ALS ALS Study Group. A placebo-controlled trial of recombinant human ciliary neurotrophic (rhCNTF) factor in amyotrophic lateral sclerosis. Ann Neurol. 1996;39:256–60. Tandan R, Bromberg MB, Forshew RN, Fries TJ, Badger GJ, Carpenter J, et al. A controlled trial of amino acid therapy in amyotrophic lateral sclerosis. I. Clinical, functional, and maximum isometric torque data. Neurology 1996; 47:1220–6. Miller RG, Moore D, Young LA, Armon C, Barohn RJ, Bromberg MB, Bryan WW, et al. Placebo-controlled trial of gabapentin in patients with amyotrophic lateral sclerosis. WALS Study Group. Western Amyotrophic Lateral Sclerosis Study Group. Neurology 1996;47:1383–8. Lacomblez L, Bensimon G, Leigh PN, Guillet P, Meininger V. Dose-ranging study of riluzole in amyotrophic lateral sclerosis. Amyotrophic Lateral Sclerosis/Riluzole Study Group II. Lancet 1996;25;347:1425–31. Lai Ec, Felice KJ, Festoff BW, Gawel MJ, Gelinas DF, Kratz R, et al., and the North America ALS/IGF-I Study Group. Effects of recombinant human insulin-like growth factor-I on progression of ALS. A placebo controlled study. Neurology 1997;49:1621–30. Mitchell JD, Wokke JHJ, Borasio GD. Recombinant human insulin-like growth factor I (rhIGF-I) for amyotrophic lateral sclerosis/motor neuron disease (Cochrane Review). In: The Cochrane Library, Issue 4, 2004. Chichester, UK: John Wiley & Sons, Ltd. Lange DJ, Murphy PL, Diamond B, Appel V, Lai EC, Younger DS, Appel SH. Selegiline is ineffective in a collaborative double-blind, placebo-controlled trial for treatment of amyotrophic lateral sclerosis. Arch Neurol. 1998;55:93–6. Borasio GD, Robberecht W, Leigh PN, Emile J, Guilloff RJ, Jerusalem F, et al., and the European ALS/IGF-I Study Group. A placebo-controlled trial of insulin-like nerve growth factor-1 in amyotrophic lateral sclerosis. Neurology 1998;51:583–6. The BDNF Study Group (Phase III). A controlled trial of recombinant methionyl human BDNF in ALS. Neurology 1999;52:1427–33. Beghi E, Chio` A, Inghilleri M, Mazzini L, Micheli A, Mora G, et al., and the Italian Amyotrophic Lateral Sclerosis Study Group. A randomized controlled trial of recombinant interferon beta-1a in ALS. Neurology 2000;54:469–74. Miller RG, Moore DH, Gelinas DF, Dronsky V, Mendoza M, Barohn RJ, et al., and the Western ALS Study Group. Phase
Clinical trials in ALS
27.
28.
29.
30.
31.
32.
33.
34. 35. 36.
37.
38.
39.
40.
41.
42.
43.
44.
III randomized trial of gabapentin in patients with amyotrophic lateral sclerosis. Neurology 2001;56:843–8. Desnuelle C, Dib M, Garrel C, Favier A. A double-blind, placebo-controlled randomized clinical trial of alphatocopherol (vitamin E) in the treatment of amyotrophic lateral sclerosis. ALS riluzole-tocopherol Study Group. Amyotroph Lateral Scler Other Motor Neuron Disord. 2001;2:9–18. Groeneveld GJ, Veldink JH, van der Tweel I, Kalmijn S, Beijer C, de Visser M, et al. A randomized sequential trial of creatine in amyotrophic lateral sclerosis. Ann Neurol. 2003;53:437–45. Cudkowicz ME, Shefner JM, Schoenfeld DA, Brown RH Jr, Johnson H, Qureshi M, et al., and the Northeast ALS Consortium. A randomized, placebo-controlled trial of topiramate in amyotrophic lateral sclerosis. Neurology 2003;61:456–64. Meiningier V, Bensimon G, Bradley WG, Brooks BR, Doillet P, Eisen AA, et al. Efficacy and safety of xaliproden in amyotrophic lateral sclerosis: results of two phase III trials. Amyotroph Lateral Scler Other Motor Neuron Disord. 2004;5:107–17. Graf M, Ecker D, Horowski R, Kramer B, Riederer P, Gerlach M, et al. on behalf of the German vitamin E/ALS study group. High dose vitamin E therapy in amyotrophic lateral sclerosis as add-on therapy to riluzole: results of a placebo-controlled double-blind study. J Neural Transm. 2004; published on-line October 27. Shefner JM, Cudkowicz E, shoenfeld D, Conrad T, Taft J, Chilton M, et al., and the NEALS Consortium. A clinical trial of creatine in ALS. Neurology 2004;63:1656–61. de Carvalho M, Scotto M, Lopes A, Swash M. Clinical and neurophysiological evaluation of progression in amyotrophic lateral sclerosis. Muscle Nerve 2003;28:630–3. de Carvalho M, Scotto M, Lopes A, Swash M. Quantifying progression in ALS. Neurology 2005;64:1783–5. Cedarbaum J. Survival. Amyotroph Lateral Scler Other Motor Neuron Disord. 2004;5(Suppl 1):79–83. Pinto AC, Evangelista T, de Carvalho M, Alves M, SalesLuı´s ML. Respiratory assistance with non-invasive ventilation (Bipap) in MND/ALS patients: survival rate in a controlled trial. J Neurol Sci. 1995;129(Suppl):19–26. Desport JC, Preux PM, Truong CT, Courat L, Vallat JM, Couratier P. Nutritional assessment and survival in ALS patients. Amyotroph Lateral Scler Other Motor Neuron Disord. 2000;1:91–6. Munsat TL, Andres PL, Finison L, Conlon T, Thibodeau L. The natural history of motor neuron loss in amyotrophic lateral sclerosis. Neurology 1988;38:409–11. Lyall RA, Donaldson N, Polkey MI, Leigh PN, Moxham J. Respiratory muscle strength and ventilatory failure in amyotrophic lateral sclerosis. Brain 2001;124:2000–13. Fitting JW, Paillex R, Hirt L, Aebischer P, Schluep M. Sniff nasal pressure: a sensitive respiratory test to assess progression of amyotrophic lateral sclerosis. Ann Neurol. 1999;46: 887–93. Norris FH Jr, Calanchini PR, Fallat RJ, Panchari S, Jewett B. The administration of guanidine in amyotrophic lateral sclerosis. Neurology 1974;24:721–8. Appel V, Stewart SS, Smith G, Appel SH. A rating scale for amyotrophic lateral sclerosis: description and preliminary experience. Ann Neurol. 1987;22:328–33. Cedarbaum JM, Stambler N. Performance of Amyotrophic Lateral Sclerosis Functional Rating Scale (ALSFRS) in multicenter clinical trials. J Neurol. 1997;152(Suppl 1): S1–9. Traynor BJ, Zhang H, Shefner JM, Schoenfeld D, Cudkowicz M, and EALS Consortium. Functional outcome measures as clinical trial endpoints in ALS. Neurology 2004;63:1933–5.
211
45. Swash M. Health outcome and quality-of-life measurements in amyotrophic lateral sclerosis. J Neurol. 1997;244 (Suppl 2):S26–9. 46. Miller RG, Moore DH. ALS trial design: expectation and reality. Amyotroph Lateral Scler Other Motor Neuron Disord. 2004;5(Suppl 1):52–4. 47. Cedarbaum JM, Stambler N, Malta E, Fuller C, Hilt D, Thurmond B, Nakanishi A. The ALS-FRS: a revised ALS functional rating scale that incorporates assessments of respiratpry function. BNDF ALS Study Group (phase II/ III). J Neurol Sci. 1999;169:13–21. 48. Swash M, de Carvalho M. The neurophysiological ´ındex in ALS. Amyotroph Lateral Scler Other Motor Neuron Disord. 2004;5(Suppl 1):10810. 49. de Carvalho M, Chio A, Dengler R, Hecht M, Weber M, Swash M. Neurophysiological measures in amyotrophic lateral sclerosis: markers of progression in clinical trials. Amyotroph Lateral Scler Other Motor Neuron Disord., in press. 50. Swash M. What does the neurologist expect from clinical neurophysiology? Muscle Nerve 2002;11(Suppl):S1348. 51. Fareed GC, Tyler R. The use of isoprinosine in patients with amyotrophic lateral sclerosis. Neurology 1971;21:937–40. 52. Olson WH, Simons JA, Halaas GW. Therapeutic trial of tilorone: lack of benefit in a double-blind, placebo-controlled study. Neurology 1978;28:1293–5. 53. Munsat TL, Easterday CS, Levy S, Wolff SM, Hiatt R. Amantadine and guanidine are ineffective in ALS. Neurology 1981;31:1054–5. 54. Bradley WG. Double-blind controlled trial of purified brain gangliosides in amyotrophic lateral sclerosis and experience with peripheral neuropathies. Adv Exp Med Biol. (1984);174:565–73. 55. Bradley WG, Hedlund W, Cooper C, Desousa GJ, Gabbai A, Mora JS, et al. A double-blind controlled trial of bovine brain gangliosides in amyotrophic lateral sclerosis. Neurology 1984;34:1079–82. 56. Harrington H, Hallett M, Tyler HR. Gangliosides therapy for amyotrophic lateral sclerosis: a double-blind controlled trial. Neurology 1984;34:1083–5. 57. Hallett M, Harrington H, Tyler HR, Flood T, Slater N. Trials of gangliosides for amyotrophic lateral sclerosis and diabetic neuropathy. Adv Exp Med Biol. 1984;174:575–9. 58. Imoto K, Saida K, Iwamura K, Saida T, Nishitani H. Amyotrophic lateral sclerosis: a double-blind crossover trial of thyrotropin-releasing hormone. J Neurol Neurosurg Psychiatry 1984;47:1332–4. 59. Olarte MR, Shafer SQ. Levamisole is ineffective in the treatment of amyotrophic lateral sclerosis. Neurology 1985;35:1063–6. 60. Mitsumoto H, Salgado ED, Negroski D, Hanson MR, Salanga VD, Wilber JF, et al. Amyotrophic lateral sclerosis: effects of acute intravenous and chronic subcutaneous administration of thyrotropin-releasing hormone in controlled trials. Neurology 1986;36:152–9. 61. Brooke MH, Florence JM, Heller SL, Kaiser KK, Phillips D, Gruber A, et al. Controlled trial of thyrotropin releasing hormone in amyotrophic lateral sclerosis. Neurology 1986;36:146–51. 62. Caroscio JT, Cohen JA, Zawodniak J, Takai V, Shapiro A, Blaustein S, et al. A double-blind, placebo-controlled trial of TRH in amyotrophic lateral sclerosis. Neurology 1986;36:141–5. 63. Norris FH, Denys EH, Fallat RJ. Trial of octaconasol in amyotrophic lateral sclerosis. Neurology 1986;36:1263–4. 64. Plaitakis A, Smith J, Mandeli J, Yahr MD. Pilot trial of branched-chain amino acids in amyotrophic lateral sclerosis. Lancet 1988;1:1015–8. 65. Lacomblez L, Bouche P, Bensimon G, Meininger V. A double-blind, placebo-controlled trial of high doses of
212
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
M. de Carvalho et al.
gangliosides in amyotrophic lateral sclerosis. Neurology 1989;39:1635–7. Blin O, Serratrice G, Pouget J, Aubrespy G, Guelton C, Crevat A. Short-term double-blind vs. placebo trial of Lthreonine in amyotrophic lateral sclerosis. Presse Med. 1989;30:1469–70. Testa D, Caraceni T, Fetoni V. Branched-chain amino acids in the treatment of amyotrophic lateral sclerosis. J. Neurol. 1989;236:445–7. Blin O, Desnuelle C, Guelton C, Aubrespy G, Ardissonne JP, Crevat A, et al. Anomalie des acides amines neurotransmetteurs dans la slerose lateral amyotrophique: une application therapeutique. Rev Neurol. 1991;147:392–4. Blin O, Pouget J, Aubrespy G, Guelton C, Crevat A, Serratrice G. A double-blind placebo-controlled trial of Lthreonine in amyotrophic lateral sclerosis. J. Neurol. 1992;239:79–81. Gil R, Neau JP, Courtois P, Gaucher C, Jonveaux T, Rosolacci T, et al. Essai en double aveugle contre placebo des acides amines branches et de la L-threonine dans le traitment des signes et symptomes de la sclerose lateral amyotrophique sur un courte periode. Sem Hop Paris 1992;68:1472–5. Testa D, Caraceni T, Fetoni V, Girotti. Chronic treatment with L-threonine in amyotrophic lateral sclerosis: a pilot study. Clin Neurol Neurosurg. 1992;94:7–9. Blin O, Pouget J, Aubrespy G, Guelton C, Crevat A, Serratrice G. A double-blind placebo-controlled trial of Lthreonine in amyotrophic lateral sclerosis. J Neurol. 1992;239:79–81. Askmark H, Aquilonius SM, Gillberg PG, Liedholm LJ, Stalberg E, Wuopio R. A pilot trial of dextromethorphan in amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry 1993;56:197–200. Hesselmans LF, Wieneke GH, Oey PL, Groenhout DM, van der Graaf Y, Gispen WH, Jennekens FG. Evaluation of electrophysiological and clinical tests in an exploratory trial of Org 2766 in motor neuron disease. Neuromuscul Disord. 1993;3:319–25. Eisen A, Stewart H, Schulzer M, Cameron D. Anti-glutamate therapy in amyotrophic lateral sclerosis. Can J Neurol Sci. 1993;20:297–301. Mazzini L, Testa D, Balzarini C, Mora G. An open.randomized clinical trial of selegiline in amyotrophic lateral sclerosis. J Neurol. 1994;241:223–7. Jossan SS, Ekblom J, Gudjonsson O, Hagbarth KE, Aquilonius SM. Double-blind crossover trial with deprenyl
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
in amyotrophic lateral sclerosis. J Neural Transm. 1994;41:237–41. Miller RG, Shepherd R, Dao H, Khramstov A, Mendonza M, Graves J, Smith S. Controlled trial of nimodipine in amyotrophic lateral sclerosis. Neuromusc Disord. 1996;6: 101–4. Blin O, Azulay JP, Desnuelle C, Bille-Turc F, Braguer D, Besse D, et al. A controlled one-year trial of dextromethorphan in amyotrophic lateral sclerosis. Clin Neuropharmacol. 1996;19:189–92. Gredal O, Werdelin L, Bak S, Christensen PB, Boysen G, Kristensen MO, et al. A clinical trial of dextromethorphan in amyotrophic lateral sclerosis. Acta Neurol Scand. 1997;96: 8–13. Kaji R, Kodama M, Imamura A, Hashida T, Kohara N, Ishizu M, et al. Effect of ultra-high dose methylcobalamin on compound muscle action potentials in amyotrophic lateral sclerosis: a double-blind controlled study. Muscle Nerve 1998;21:1775–8. Chio A, Cucatto A, Terreni AA, Schiffer D. Reduced glutathione in amyotrophic lateral sclerosis: an open, crossover, randomized trial. Ital J Neurol Sci. 1998;19:363–6. Bensimon G, Lacomblez L, Delumeau JC, Bejuit R, Truffinet P, Meininger V. Riluzole/ALS Study Group. A study of riluzole in the treatment of advanced stage or elderly patients with amyotrophic lateral sclerosis. Journal of Neurology 2002;249:609–15. Ryberg H, Askmark H, Persson LI. A double-blind randomized clinical trial in amyotrophic lateral sclerosis using lamotrigine: effects on CSF glutamate, aspartate, branchedchain amino acid levels and clinical parameters. Acta Neurol Scand. 2003;108:1–8. Yuen EC, Olney RK. Longitudinal study of fibre density and motor unit number estimate in patients with amyotrophic lateral sclerosis. Neurology 1997;49:573–8. Felice KJ. A longitudinal study comparing thenar motor unit number estimates to other quantitative tests in patients with amyotrophic lateral sclerosis. Muscle Nerve 1997;20: 179–85. Bromberg MB, Fries TJ, Forshew DA, Tandan R. Electrophysiological endpoint measures in a multicentre ALS drug trial. J Neurol Sci. 2001;184:51–5. Shefner JM, Cudkowicz ME, Zhang H, Schoenfeld D, Jillapalli D, Northeast ALS Consortium. The use of statistical MUNE in a multicentre clinical trial. Muscle Nerve 2004;30:463–9.
