You are here

Article Text

Article menu

Download PDFPDF

Neuromuscular
Research paper
Rate of disease progression: a prognostic biomarker in ALS
Free
  1. Julie Labra1,
  2. Parvathi Menon2,
  3. Karen Byth3,4,
  4. Shea Morrison1,
  5. Steve Vucic2
  1. 1St Joseph's Hospital, Sydney, Australia
  2. 2Western Clinical School, University of Sydney, Sydney, Australia
  3. 3Westmead Hospital, Research and Education Network, Sydney, Australia
  4. 4NHMRC Clinical Trials Centre, University of Sydney, Sydney, Australia
  1. Correspondence to Professor Steve Vucic, Department of Neurology, Westmead Hospital, Cnr Hawkesbury and Darcy Roads, Westmead, NSW 2145, Australia; s.vucic{at}neura.edu.au

Abstract

Objective To assess the utility of rate of disease progression (ΔFS) as a prognostic biomarker in amyotrophic laterals sclerosis (ALS).

Methods A total of 203 patients with ALS were prospectively recruited over a 10-year period. At initial visit, the following variables were collected: demographic details, symptom duration, site of onset, phenotype, riluzole use and Amyotrophic Lateral Sclerosis Functional Rating Scale-Revised (ALSFRS-R) scores. The ΔFS score at initial visit was calculated by dividing the ALSFRS-R total score by symptom duration (months). The primary end point was survival. Kaplan-Meier survival curves were used to illustrate the distribution of survival from a specified point, while multiple Cox proportional hazards modelling with backward stepwise variable selection was used to identify the independent predictors of survival at initial visit.

Results The ΔFS score at initial visit was a significant predictor of survival in ALS (p<0.001), and remained significant when adjusted for age and site of onset (p<0.001). 3 prognostic subgroups emerged, with a ΔFS score of <0.47 associated with a median survival of 2.4 years, which was significantly greater when compared with an initial ΔFS score of between 0.47 and 1.11 (1.6 years, p<0.05) and a score >1.11 (0.7 years, p<0.001). Importantly, multiple Cox proportional hazards modelling identified ΔFS as a highly significant independent predictor of survival in ALS (p<0.001) along with site of disease onset (p<0.01).

Conclusions Rate of disease progression appears to be a simple and sensitive clinical prognostic biomarker in ALS that could be potentially utilised in clinical practice and future therapeutic trials.

http://dx.doi.org/10.1136/jnnp-2015-310998

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    The development of clinically applicable and reliable prognostic biomarkers in amyotrophic lateral sclerosis (ALS) would be of vital importance for guiding patient management and recruitment into clinical treatment trials.1 In a clinical setting, the availability of useful prognostic biomarkers would enable implementation of an appropriate care plan.2 From a research perspective, the identification of sensitive and clinically applicable biomarkers would enable equal stratification of patients into treatment groups (active vs placebo) based on the rate of disease progression, and provide an opportunity to promptly evaluate drug effectiveness.

    A number of demographic biomarkers including older age, bulbar-onset disease, respiratory dysfunction and nutritional status have been correlated with adverse prognosis in ALS.2–5 The limitation in utilising these clinical biomarkers pertains to heterogeneity of the ALS phenotype, particularly in determining the differences in disease progression rates within patients.6 Of further relevance, blood, cerebrospinal fluid, neurophysiological and neuroradiological biomarkers have also been reported, although the reliability and clinical applicability of such biomarkers in ALS remains limited.1

    The Amyotrophic Lateral Sclerosis Functional Rating Scale-Revised (ALSFRS-R) has been reported as a prognostic biomarker predictor of survival in ALS.7 ,8 A potential limitation of total ALSFRS-R score is the failure to incorporate symptom duration, an independent predictor of survival.9 ,10 Consequently, the rate of disease progression (ΔFS) score was developed which expressed the ALSFRS-R as a function of the disease duration.11 A study in a Japanese ALS cohort suggested that the ΔFS at initial assessment was an independent and reliable biomarker of disease progression in individual patients when compared with the ALSFRS-R,11 a finding confirmed in a subsequent study.12 The prognostic utility of ΔFS requires further validation, particularly the development of statistically derived prognostic cut-off values, prior to being accepted as a useful prognostic biomarker. Consequently, the present study assessed the utility of ΔFS as a prognostic clinical biomarker in a large Australian ALS cohort, with the aim of developing reliable prognostic groups.

    Methods

    Patients

    The ALS cohort was prospectively recruited between 1 June 2004 and 31 July 2014, and registered on the St Joseph's Hospital ALS database, a contributor to the Australian motor neuron disease registry (AMNDR), the largest clinical ALS database in Australia.13 The cohort in the ALS database comprised 203 patients with ALS. All patients were clinically assessed in the multidisciplinary ALS clinic every 3–6 months, depending on clinical need, and were followed up until death or census date (31 July 2014). The primary end point was survival from initial visit, calculated by time to death or disease duration at census date. This study was approved by the Sydney West Area Health Service and the St Vincent's Hospital Sydney Human Research Ethics Committees.

    Data was collected for the following variables: gender, age at initial visit (years), symptom onset date (obtained from patient or family reports) and disease duration at initial visit in clinic, diagnosis date, phenotype, ALSFRS-R total score,7 percutaneous endoscopic gastrostomy (PEG) tube insertion and institution of non-invasive ventilation. The rate of disease progression (ΔFS) was calculated at the initial visit using the following formula:11Embedded Image

    Statistical analysis

    All statistical analyses were performed using the SPSS Statistics package (V.21), and the analyses were exploratory. Two-tailed tests with a significance level of 5% were used throughout. Continuous variables were often skewed and are summarised using the median and lower quartile (LQ) to upper quartile (UQ). The χ2 or exact permutation tests were used as appropriate to test for association between categorical variables. Kruskal-Wallis non-parametric analysis of variance was used to assess differences between means. Kaplan-Meier survival curves were used to illustrate the distribution of survival from a specified point, and log rank tests were used to test for differences between groups in the survival from initial visit. To facilitate these analyses, potential risk factors with continuous distributions were split into categories as follows: age (<70, ≥70 years), ΔFS tertiles (<0.47, 0.47–1.11, >1.11), ALSFRS-R quartiles (≤33, 34–39, 40–43, ≥44), ALSFRS-R respiratory subscore tertiles (<10, 10–11, 12) and symptom duration quartiles (<0.7, 0.7–1, 1.1–2, >2 years). Univariable analysis was initially performed to assess which covariates were associated with survival. All potential risk factors demonstrating a univariable association with survival from initial visit were considered as candidate variables for inclusion in a multiple Cox proportional hazards model with backward stepwise variable selection, in order to identify the independent predictors of survival from initial visit. HR with 95% CIs were used to quantify the degree of association between a risk factor and survival from initial visit in clinic.

    Results

    The study was undertaken on 203 patients with ALS (men 111, women 92, median age 63 years, age range 24–89 years) diagnosed in accordance with the revised El Escorial criteria.14 Seventy-two per cent of patients were classified in the El Escorial definite/probable diagnostic categories, while 28% were classified as possible ALS. Thirty-six patients were excluded from the study due to an incomplete data set, and these patients were not clinically different from the ALS cohort recruited into the study. Three patients were excluded due to the fact that a non-ALS neuromuscular disorder (two with Kennedy's disease and one with autoimmune motor neuropathy) was established during follow-up. Consequently, the cohort for this study comprised 164 registered patients with ALS (93 men, 71 women, median age 65 years, range 26–89 years) and their clinical features are summarised in table 1.

    View this table:
    Table 1

    Clinical characteristics for patients with amyotrophic lateral sclerosis (ALS) as a group and divided into three tertiles as defined by the rate of disease progression (ΔFS)

    The median age of patients at initial visit was 64.7 years (58–73 years), with the median disease duration from symptom onset to assessment in clinic being 12 months (7.2, 24 months). In order to ensure that the timing of symptom onset was accurate, information was also sought from patients’ relatives. Limb-onset disease was evident in 58% of patients with ALS, while bulbar-onset disease was reported in 42% patients. At initial assessment, the median ALSFRS-R score was 39 (33–43), while the median ΔFS score was 0.67 (0.33–1.47). Seventy-seven per cent of patients were receiving riluzole at or during the follow-up period. During the follow-up period, 49% (N=81) of patients underwent a PEG insertion at a median time of 17.5 months (12–28 months) after symptom onset, while 21% (N=34) of patients were started on non-invasive ventilation at a median time of 20.0 months (12–36 months) after symptom onset. None of the patients had tracheostomy-assisted ventilation.

    In total, 77% of patients with ALS had died by the census date (31 July 2014). Utilising Kaplan-Meier plots, the median survival from symptom onset was estimated to be 2.7 years (95% CI 2.3 to 3.1 years), while the median survival time from diagnosis of ALS was 1.7 years (95% CI 1.5 to 2.0 years), and that from initial visit to the clinic 1.4 years (95% CI 1.2 to 1.6 years). In addition, the percentage of patients surviving at 1, 2, 3 and 5 years after initial assessment in the clinic was 59±3.9%, 33±3.8%, 23±3.6%, and 21±3.5%, respectively, and is in keeping with previous studies.3 ,10 ,15

    Prognostic biomarkers in ALS

    In order to determine whether ΔFS was a prognostic biomarker of ALS at initial assessment, a univariable analysis was performed. The rate of disease progression at initial assessment was a significant predictor of survival in patients with ALS when assessed at initial visit (p<0.001, table 2). The association remained highly significant when adjusted for age and site of onset (p<0.001). Specifically, a ΔFS score of <0.47 at initial assessment was associated with a median survival of 2.4 years, which was significantly greater when compared with an initial ΔFS score of between 0.47 and 1.11 (1.6 years, HR 1.77, 95% CI 1.12 to 2.82, p<0.05) and a score >1.11 (0.7 years, HR 3.74, CI 2.40 to 5.82, p<0.001, figure 1, table 2).

    View this table:
    Table 2

    Univariable survival analysis disclosed that a higher rate of disease progression (ΔFS), along with lower Amyotrophic Laterals Sclerosis Rating Scale-Revised (ALSFRS-R) score, older age of onset (≥70 years), bulbar onset disease and shorter disease duration were predictors of survival

    Figure 1
    Figure 1

    Kaplan-Meier survival curves indicating that a rate of disease progression (ΔFS) <1.11 at initial assessment was associated with poorest survival.

    Subgroup analysis of the three ΔFS groups disclosed a greater proportion of patients with bulbar onset disease in the highest tertile group (ΔFS>1.11) as well as a greater frequency of the classical ALS phenotype (table 1). In addition, patients in the highest tertile groups were significantly older (≥70 years), and exhibited a significantly shorter symptom duration and time to definite/probable ALS diagnosis (table 1). Of further relevance, the total ALSFRS-R score was significantly smaller (table 2), as were the ALSFRS subscores (figure 2). Importantly, the proportion of patients receiving riluzole in the highest tertile group was not significantly different from more favourable tertile groups (table 1).

    Figure 2
    Figure 2

    The median Amyotrophic Laterals Sclerosis Rating Scale-Revised (ALSFRS-R) subscores were significantly smaller in patients with the highest rate of disease progression (ΔFS; ***p<0.001; ****p<0.0001).

    In addition to the rate of disease progression score, older age ≥70 years, shorter duration of symptoms, lower total ALSFRS R score, ALSFRS-R respiratory subscore <10, and bulbar onset disease at initial visit were also identified as adverse prognostic biomarkers in ALS on univariable analysis (table 2). By contrast, gender, therapy with riluzole, presence of percutaneous endoscopic gastrostomy (PEG) tube or non invasive ventilation (NIV) use were not significant predictors of survival in the present cohort (table 2).

    Since several variables were associated with survival from initial visit on univariable analysis, multiple Cox proportional hazards modelling was used to explore the relationship between survival in ALS and six candidate variables: age at initial visit (<70 vs ≥70 years), site of onset (limb vs bulbar), symptom duration at initial visit quartiles (years), ALSFRS-R total score, ALSFRS-R respiratory subscore and ΔFS tertiles. The best-fitting Cox proportional hazards model obtained using backward stepwise variable selection, identified ΔFS tertiles at initial assessment and the site of disease onset as independent significant predictors of survival in ALS (table 3).

    View this table:
    Table 3

    Best fitting Cox proportional hazards model showing independent predictors of survival at initial visit

    Discussion

    The findings in the present study have further validated the prognostic utility of the ΔFS score at initial assessment in an Australian ALS cohort, establishing ΔFS as an independent and robust prognostic biomarker for ALS. In addition, the three statistically derived prognostic groups emerged (ΔFS<0.47, 0.47–1.11 and >1.11), not previously reported, could be potentially utilised as prognostic guides in patients management and future clinical trials. Importantly, a ΔFS score of >1.11 resulted in survival times of only 8.4 months and a 3.7-fold greater risk of death, when compared with patients with a ΔFS score <0.47. In addition, older age (≥70 years), shorter disease duration (≤2 years), lower total ALSFRS-R score (≤33), ALSFRS-R respiratory subscore <10 and bulbar-onset disease were also adverse prognostic factors on univariable analysis. Importantly, multiple Cox proportional hazard modelling identified ΔFS at initial clinic visit and bulbar-onset disease as independent predictors of survival in ALS. These results suggest a potential utility of ΔFS as a prognostic biomarker in ALS, as it robustly discriminates between three prognostic subgroups.

    Prognostic biomarkers in ALS

    The development of reliable and sensitive prognostic biomarkers for ALS is of vital importance in patient management and stratification into clinical trials.1 Specifically, identifying patients with aggressive disease at early stages of ALS may enable more adequate patient care planning, including timing of institution of communication devices, gastrostomy feeding and non-invasive ventilation.16 In addition, prognostic biomarkers may increase the power of clinical trials by reducing subject heterogeneity,17 and could be utilised as primary end points, assessing the pharmacodynamic effect of drugs in early stages of development. Ideally, a prognostic biomarker should reliably discriminate between ALS subgroups, and be easily accessible and practical to implement.1

    The findings in the present study establish ΔFS at initial clinic visit, as a sensitive prognostic biomarker in ALS. Specifically, the ΔFS robustly discriminates between three different prognostic ALS subgroups, as indicated by multiple Cox proportional hazard models where ΔFS was an independent predictor of survival. In addition, ΔFS is simple and inexpensive to administer, requiring disease duration and the ALSFRS-R score, both of which are routinely clinically collated. Importantly, the present findings are in keeping with previous studies which have identified ΔFS as a prognostic biomarker for ALS survival at initial assessment.11 ,12

    The main limitation in utilising the ΔFS as a prognostic biomarker in ALS pertains to an absence of universally accepted prognostic values. Previous studies have assessed the utility of ΔFS by dividing patients into two groups about a median ΔFS value, and have arbitrarily assigned three prognostic cut-off values (<0.5; 0.5–1 and ≥1).5 ,9 In addition, it has also been suggested that ΔFS scores may exhibit a regional variation, with the suggestion that centres should develop unique ΔFS score cut-off values, thereby potentially limiting the ΔFS score as a ubiquitous prognostic biomarker in ALS. It is reassuring that the median ΔFS score of 0.67 in the present study was similar to previous findings,11 as were the three statistically derived prognostic cut-off values (<0.47, 0.47–1.11, >1.11). As such, the ΔFS score prognostic values appear to be similar irrespective of the origins of the ALS cohorts, thereby supporting the utility of the ΔFS score as a prognostic biomarker in ALS.

    In addition to the ΔFS score, age ≥70 years, symptom duration ≤2 years, ALSFRS-R total score ≤33, ALSFRS-R respiratory subscore <10, and bulbar-onset disease were also established as potential adverse prognostic biomarkers in the current ALS cohort, albeit on univariable analysis. Interestingly, while there was a trend for riluzole therapy to increase survival in ALS, this was not significant in the current cohort and probably relates to sample size, with the majority of patients with ALS (77%) receiving riluzole therapy. Importantly, multiple Cox proportional hazards modelling established that in addition to the ΔFS, site of disease onset was the only other independent prognostic biomarker of survival in ALS. While the site of disease onset could be utilised in patient care planning and stratification into clinical trials, the utility of site of disease onset as a biomarker of treatment efficacy appears limited when compared with the ΔFS score, as ΔFS is a numerical parameter that could be modulated by therapies, a notion that needs verification in future studies. It should be stressed, however, that the ΔFS score would not necessarily determine the need for early intervention, such as insertion of PEG or the start of NIV, and that such decisions remain dependent on individual patient assessment and are likely to be determined by a number of factors.

    A potential limitation of the present study is the absence of quantitative respiratory function data, including the forced vital capacity. Instead, we utilised the ALSFRS-R respiratory subscore as a measure of respiratory function at initial visit. While the presence of respiratory dysfunction adversely impacts on survival,18 previous studies did not establish a significant difference in the forced vital capacity between ‘faster’ and ‘slower’ progressing patients.11 Of further relevance, while the median ALSFRS-R respiratory subscore was significantly lower in patients with the fastest rate of disease progression (ΔFS>1.11), it was found not to be an independent predictor of survival on multiple Cox proportional hazards analysis. Taken together, the present findings suggest that while patients with ALS with aggressive disease (ΔFS>1.11) exhibit respiratory dysfunction at initial visit, significant respiratory dysfunction is unlikely, and typically manifests in later stages of the disease process.

    It has also been argued that a prolonged presymptomatic period may be a feature of ALS, akin to that seen in other neurodegenerative disorders, with the pathogenic processes culminating in the ALS phenotype starting much earlier in life, perhaps at the time of conception.19 By contrast, others have suggested that the pathogenic processes develop just before or co-incidental with the clinical onset of ALS.20 ,21 Irrespective of the length of the ‘preclinical period’, the findings in the present study suggest that once the disease starts, the ΔFS score appears to be a useful prognostic biomarker.

    In conclusion, the present study established the ΔFS score as a sensitive and independent prognostic biomarker for ALS. Specifically, three prognostic ΔFS subgroups emerged, all correlating with survival times. Importantly, the cut-off values for the prognostic subgroups were similar to the previously reported values which were arbitrarily derived from distinct ALS cohorts, supporting the utility of ΔFS score as a ubiquitous prognostic biomarker in ALS. Future studies should assess whether the ΔFS could be utilised as a biomarker of therapeutic effectiveness in ALS at early stages of drug development.

    References

    1. ↵
      1. Turner MR,
      2. Kiernan MC,
      3. Leigh PN, et al
      . Biomarkers in amyotrophic lateral sclerosis. Lancet Neurol 2009;8:94–109. doi:10.1016/S1474-4422(08)70293-X
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Kiernan MC,
      2. Vucic S,
      3. Cheah BC, et al
      . Amyotrophic lateral sclerosis. Lancet 2011;377:942–55. doi:10.1016/S0140-6736(10)61156-7
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Chio A
      . ISIS Survey: an international study on the diagnostic process and its implications in amyotrophic lateral sclerosis. J Neurol 1999;246(Suppl 3):1–5. doi:10.1007/BF03161081
      OpenUrlCrossRef
    4. ↵
      1. Mandrioli J,
      2. Faglioni P,
      3. Nichelli P, et al
      . Amyotrophic lateral sclerosis: prognostic indicators of survival. Amyotroph Lat Scler 2006;7:217–26. doi:10.1080/17482960600947648
      OpenUrlCrossRef
    5. ↵
      1. Desport JC,
      2. Torny F,
      3. Lacoste M, et al
      . Hypermetabolism in ALS: correlations with clinical and paraclinical parameters. Neurodegener Dis 2005;2:202–7. doi:10.1159/000089626
      OpenUrlCrossRefPubMed
    6. ↵
      1. Lee JR-J,
      2. Annegers JF,
      3. Appel SH
      . Prognosis of amyotrophic lateral sclerosis and the effect of referral selection. J Neurol Sci 1995;132:207–15. doi:10.1016/0022-510X(95)00154-T
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Cedarbaum JM,
      2. Stambler N,
      3. Malta E, et al
      . The ALSFRS-R: a revised ALS functional rating scale that incorporates assessments of respiratory function. BDNF ALS Study Group (Phase III). J Neurol Sci 1999;169:13–21. doi:10.1016/S0022-510X(99)00210-5
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Kollewe K,
      2. Mauss U,
      3. Krampfl K, et al
      . ALSFRS-R score and its ratio: a useful predictor for ALS-progression. J Neurol Sci 2008;275:69–73. doi:10.1016/j.jns.2008.07.016
      OpenUrlCrossRefPubMed
    9. ↵
      1. Gordon PH,
      2. Salachas F,
      3. Lacomblez L, et al
      . Predicting survival of patients with amyotrophic lateral sclerosis at presentation: a 15-year experience. Neurodegener Dis 2013;12:81–90. doi:10.1159/000341316
      OpenUrlCrossRefPubMed
    10. ↵
      1. Zoccolella S,
      2. Beghi E,
      3. Palagano G, et al
      . Analysis of survival and prognostic factors in amyotrophic lateral sclerosis: a population based study. J Neurol Neurosurg Psychiatry 2008;79:33–7. doi:10.1136/jnnp.2007.118018
      OpenUrlAbstract/FREE Full Text
    11. ↵
      1. Kimura F,
      2. Fujimura C,
      3. Ishida S, et al
      . Progression rate of ALSFRS-R at time of diagnosis predicts survival time in ALS. Neurology 2006;66:265–7. doi:10.1212/01.wnl.0000194316.91908.8a
      OpenUrlCrossRefPubMed
    12. ↵
      1. Gordon PH,
      2. Cheung YK
      . Progression rate of ALSFRS-R at time of diagnosis predicts survival time in ALS. Neurology 2006;67:1314–15. doi:10.1212/01.wnl.0000243812.25517.87
      OpenUrlCrossRefPubMed
    13. ↵
      1. Kiernan MC,
      2. Talman P,
      3. Henderson RD, et al
      . Establishment of an Australian motor neurone disease registry. Med J Aust 2006;184:367–8.
      OpenUrlPubMedWeb of Science
    14. ↵
      1. Brooks B,
      2. Miller R,
      3. Swash M, et al
      . El Escorial revisited: revised criteria for the diagnosis of amyotrophic lateral sclerosis. Amyotroph Lateral Scler Other Motor Neuron Disord 2000;1:293–9. doi:10.1080/146608200300079536
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Traynor BJ,
      2. Alexander M,
      3. Corr B, et al
      . Effect of a multidisciplinary amyotrophic lateral sclerosis (ALS) clinic on ALS survival: a population based study, 1996–2000. J Neurol Neurosurg Psychiatry 2003;74:1258–61. doi:10.1136/jnnp.74.9.1258
      OpenUrlAbstract/FREE Full Text
    16. ↵
      1. Vucic S,
      2. Rothstein JD,
      3. Kiernan MC
      . Advances in treating amyotrophic lateral sclerosis: insights from pathophysiological studies. Trends Neurosci 2014;37:433–42. doi:10.1016/j.tins.2014.05.006
      OpenUrlCrossRefPubMed
    17. ↵
      1. Nicholson K,
      2. Cudkowicz M,
      3. Berry J
      . Clinical trial designs in amyotrophic lateral sclerosis: does one design fit all? Neurotherapeutics 2015;12:376–83. doi:10.1007/s13311-015-0341-2
      OpenUrlCrossRefPubMed
    18. ↵
      1. Czaplinski A,
      2. Yen AA,
      3. Appel SH
      . Forced vital capacity (FVC) as an indicator of survival and disease progression in an ALS clinic population. J Neurol Neurosurg Psychiatry 2006;77:390–2. doi:10.1136/jnnp.2005.072660
      OpenUrlAbstract/FREE Full Text
    19. ↵
      1. Eisen A,
      2. Kiernan M,
      3. Mitsumoto H, et al
      . Amyotrophic lateral sclerosis: a long preclinical period? J Neurol Neurosurg Psychiatry 2014;85:1232–8. doi:10.1136/jnnp-2013-307135
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Aggarwal A,
      2. Nicholson G
      . Detection of preclinical motor neurone loss in SOD1 mutation carriers using motor unit number estimation. J Neurol Neurosurg Psychiatry 2002;73:199–201. doi:10.1136/jnnp.73.2.199
      OpenUrlAbstract/FREE Full Text
    21. ↵
      1. de Carvalho M,
      2. Swash M
      . The onset of amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry 2006;77:388–9. doi:10.1136/jnnp.2005.073031
      OpenUrlAbstract/FREE Full Text

    Footnotes

    • Funding Funding support from the Motor Neuron Disease Research Institute of Australia (MNDRIA) and Medical Research Council of Australia (Project grant number APP1055778) is gratefully acknowledged.

    • Competing interests None declared.

    • Patient consent Obtained.

    • Ethics approval This study was approved by the Sydney West Area Health Service and the St Vincent's Hospital Sydney Human Research Ethics Committees’.

    • Provenance and peer review Not commissioned; externally peer reviewed.