P92 DETECTION OF ALS BY QUANTITATIVE REAL TIME RT‐PCR OF PERIPHERAL BLOOD CELLS
Yu S1, Do H1, Mass J2, Katz J2, Zhang R3, Gascon R3, Lancero M1, Miller RG2, McGrath MS3 & Hadlock KG1
1Pathologica LLC, Burlingame, 2Forbes Norris MDA/ALS Research Center, San Francisco, and 3University of California, San Francisco, San Francisco, USA
E‐mail address for correspondence: [email protected]
Background: Recent work by our group has established that blood macrophages from individuals with amyotrophic lateral sclerosis (ALS) have significant changes in their transcriptional program compared to healthy individuals. This raises the possibility that diagnosis of ALS can be assisted by molecular analysis of blood specimens.
Objective: To evaluate the utility of a quantitative RT‐PCR assay of peripheral blood cells for the classification of samples as derived from ALS patients or controls.
Methods: Blood samples were obtained from patients with ALS and controls after obtaining informed consent and mononuclear cells were isolated by Percoll gradient centrifugation. The cells were cultured overnight at 37°C under non‐adherent conditions, collected, lysed, and total RNA prepared. Aliquots of the RNA preparations were then RT‐PCR amplified with primer pairs specific for five human genes using a Light‐Cycler. The signals obtained were normalized against the signal obtained with primers for β‐actin. A classification algorithm was built according to published methods (1) using signals obtained from 13 randomly selected ALS patients and 11 healthy controls. The algorithm was then tested using additional samples derived from ALS patients (n = 18), healthy controls (n = 9), and individuals with macular degeneration (MDgn) (MDgn, n = 16). Operators of the Light‐Cycler were blinded with respect to the classification algorithm. Individuals with MDgn were evaluated as controls because they also have activated macrophages in their blood.
Results: In the training set 12 of 13 samples from ALS patients and 0 of 11 control samples were classified as ALS‐like. In the test samples 17 of 18 ALS samples, 0 of 9 control samples, and 0 of 16 MDgn samples were classified as ALS‐like. All eight samples obtained from ALS patients within six months of diagnosis were classified as ALS‐like and there was no obvious trend towards higher or lower scores with advanced disease; nor was use of riluzole or other medications by ALS patients associated with significant changes in classification scores.
Conclusions: A classification system based on the transcription level of five genes expressed in peripheral blood macrophages of ALS patients exhibited ∼95% sensitivity (29 of 31 samples/24 of 25 patients) and 100% specificity (0 of 36 controls) at identifying ALS patients. Thus diagnostics based on alterations in gene expression seen in peripheral blood macrophages hold great promise for the early detection of ALS. Ongoing studies are focused on evaluating the classification index in patients with other neurological diseases.
Reference
- Golub TR, Slonim DK, Tamayo P, et al. Science 1999;286:531–7.
P93 CIRCULATING ANTIBODIES TO CYSTEINYL CATECHOLAMINES IN AMYOTROPHIC LATERAL SCLEROSIS PATIENTS
Salauze L1, van der Velden C2, Poulletier de Gannes F1, Lagroye I1, Veyret B1 & Geffard M1
1PIOM laboratory, PESSAC, and 2Gemacbio, Cenon, France
E‐mail address for correspondence: [email protected]
Background: Recent studies, so far not related to motor neurons, have shown in vivo formation of cysteinyl‐catecholamine (Cyst‐CA) compounds, which may have a neurotoxic activity. These compounds result from the oxidation of catecholamines followed by the nucleophilic reaction with thiol residues such as cysteine.
Objectives: Using a routine ELISA methodology, circulating antibodies (IgG, IgM and IgA) to Cyst‐CA were searched in ALS patients' sera.
Methods: All sera were obtained with the patients' informed consent. Two groups of patients were studied: 1) 27 ALS patients (40–80 years old) affected at various sites of onset, used for preliminary screening; 2) another, larger group of 49 patients (40–80 years old) classified according to onset of the disease, including sera from 17 patients with a bulbar onset (10 males and 7 females), sera from 14 patients with an upper onset (10 males and 4 females), and sera from 18 patients with a lower onset (15 males and 3 females). These groups were being treated with riluzole®. The control and comparative populations in our studies were 1) healthy controls (n = 19, 37–70 years old); 2) patients with Parkinson's disease (n = 23, 40–80 years old) and patients with multiple sclerosis (n = 25, 40–80 years old). We synthesized Cyst‐CA compounds, mimicking their endogenous formation by linking N‐acetyl‐cysteine (NAC) with catecholamines: dopamine (DA), L‐3,4‐dihydroxyphenylalanine (L‐DOPA), adrenaline (A), noradrenaline (NA), or homovanillic acid (HVA). The different groups of patients were statistically compared to the corresponding control groups using the Mann‐Whitney U‐ test.
Results: In a first series of experiments, the antibody titres were compared to those of controls and patients with other neurodegenerative diseases. Significant antibody levels were found for Cyst‐CA. The G and A isotypes were found in contrast to the M isotype. A second series of experiments showed that A and G titres were elevated, depending on the type of Cyst‐CA and the onset of the disease. IgG for NAC‐ L‐DOPA were present in cases of bulbar and upper onsets. IgA for NAC‐HVA, NAC‐A, and NAC‐DA were found in lower onset.
Conclusions: For the first time, we have found circulating antibodies, in the sera of ALS patients, directed against Cyst‐CA compounds of IgG and IgA isotypes. No IgM antibodies were found.
These results indirectly show that 1) the oxidation of CA and the formation of Cyst‐CA may be involved in ALS; 2) these radical processes have different targets depending on the onset of the disease. These findings show that the extent, as well as the variety of radical processes in ALS is larger than expected.
P94 PROTEOMICS ANALYSIS OF MOTOR NEURON DISEASE PATIENTS' CEREBROSPINAL FLUID
Conti A, Iannaccone S, Olivieri S, Sferrazza B & Alessio M
San Raffaele Scientific Institute, Milan, Italy
E‐mail address for correspondence: [email protected]
MND is a severe fatal syndrome characterized by the progressive neurodegeneration of motor neurons. Familial and sporadic forms of the disease have indistinguishable clinical and histopathological features suggesting common molecular pathogenic mechanisms. Several hypotheses have been formulated; however, the disease mechanisms are largely unknown. One of the limitations in MND is the difficulty of diagnosis in the early stages of the disease, due to the absence of specific markers.
For this reason and to investigate pathological mechanisms we analysed the patients' CSF; this fluid surrounds the brain and spinal cord and it may be expected to reflect some changes occurring in the central nervous system, probably involved with the progression of the disease.
This analysis has been performed using the proteomic approach, a technique that allows the separation of a large number of proteins. The term refers to the study of the complete set of proteins that are expressed by the genome of a cell in a defined physiological or pathological condition. This approach includes the generation of two‐dimensional gel electrophoresis (2DE) stained maps that resolve proteins on the basis of the isoelectric point and relative mass. It is combined with protein identification by mass spectrometry (MS) and comparisons with reference to 2DE maps of many tissues, fluids, and cell lines in different conditions. By comparing proteins coming from healthy donor versus MND patients we aim to identify differentially expressed proteins produced by pathological mechanisms. Initial experiments show a protein isoform expression reduced in MND patient cerebrospinal fluid versus healthy controls.
P95 CEREBROSPINAL FLUID OXIDIZED PROTEINS AND ANTIOXIDANT CONTENT ARE DIFFERENTIALLY MODIFIED IN THE DIFFERENT FORMS OF AMYOTROPHIC LATERAL SCLEROSIS
Siciliano G1, Del Corona A1, Piazza S1, Volpi L1, Franzini M2, Paolicchi A2 & Pompella A2
1Department of Neuroscience, University of Pisa, Pisa, and 2Department of Experimental Pathology, University of Pisa, Pisa, Italy
E‐mail address for correspondence: [email protected]
Background: Amyotrophic lateral sclerosis (ALS) is a neurological disorder characterized by a progressive upper and lower motor neuron loss, usually beginning in late adulthood or in senescence and with a mean survival time not longer than three years from the diagnosis. ALS causes are unknown, but accumulating evidence including the occurrence of mutations in superoxide dismutase 1 (SOD1) gene in 20% of familial cases, indicate that oxidative stress is involved in the pathogenesis of this disease. In fact, an imbalance in the equilibrium between oxidative stress and endogenous antioxidant capacity can be responsible of damage to proteins, lipids, and DNA and can contribute to motor neuron degeneration.
Objectives: The aim of this study was to test the hypothesis that oxidative stress contributes to the pathogenesis of ALS. To do this we chose to assess 1) a generic marker of oxidative stress, i.e. the overall content of antioxidants in the cerebrospinal fluid (CSF), determined by its ferric reducing ability (FRA) and 2) a specific index of oxidative damage to proteins by reactive oxygen species, i.e. the amount of advanced oxidation protein products (AOPP).
Methods: Samples of CSF were obtained from 49 consecutive ALS patients (10 presenting with the bulbar form, 17 with the pseudopolyneuropathic form, 22 with the classical form of the disease), mean age±SD, 62.73±10.82 years and eight control subjects, mean age 41.25±14.64 years, with neurological diseases different from ALS and absence of CSF biochemical anomalies at standard laboratory analysis.
Results: AOPP results were higher in ALS patients (mean±SE, 13.2±1.7) compared to control group (0.75±0.55), p = 0.0039. However, when subdividing patients, those with the bulbar form had AOPP levels significantly lower (p<0.001) than patients with the classical or the pseudopolyneuropathic form (1.1±0.59, 15.35±2.47, and 18.02±2.57, respectively). FRA was found to be lower in patients with respect to controls (48.8±2.2 and 61.3±3.2, respectively, p<0.03), but no difference was found between the different forms of ALS.
Conclusions: ALS has been found to be associated with a decrease of CSF antioxidant ability as assessed by FRA, and increase of oxidative damaged to proteins, as assessed by AOPP. The fact that AOPP levels, but not FRA, are differently affected in the three forms of ALS suggests that although oxidative stress is involved in the pathogenesis of ALS, the precise mechanisms and nature of the oxidative challenge to CNS might be specific for each form.
P96 IDENTIFICATION OF CEREBROSPINAL FLUID POTENTIAL BIOMARKERS OF AMYOTROPHIC LATERAL SCLEROSIS
Pasinetti GM1, Ungar LH3, Lange DJ2, Yemul S1, Brown RH4, Cudkowicz ME4, Peskind E5, Marcus S6 & Ho L1
1Department of Psychiatry & 2MDA/ALS Program, Department of Neurology, Mount Sinai School of Medicine, New York, 3Department of Computation and Informatics Sciences, University of Pennsylvania, Philadelphia, 4The Day Laboratory for Neuromuscular Research, Massachusetts General Hospital, Boston, 5University of Washington School of Medicine, Department of Psychiatry and Behavioral Sciences, and 6VA Mental Illness Research, Education, and Clinical Center Harborview Medical Center, Seattle, USA
E‐mail address for correspondence: [email protected]
Background: High‐throughput protein expression analysis provides rapid and comprehensive screening for proteins in biological fluids and therefore is potentially useful for the identification of novel protein biomarkers. Amyotrophic lateral sclerosis (ALS) is a disease with a diagnosis based entirely on clinical features. Identification of a true biomarker for ALS is important for diagnosis and may provide insight into the pathogenesis of this disease.
Objective: To determine if there is a specific protein profile in the cerebrospinal fluid (CSF) that separates patients with ALS from those with pure motor and peripheral neuropathy (PN) and normal healthy control subjects.
Methods: CSF from normal healthy controls, ALS and neurological control cases followed at Mount Sinai School of Medicine (MSSM), Massachusetts General Hospital (MGH) and University of Washington (UW) were analyzed by surface enhanced laser desorption/ionization (SELDI)‐time of flight (TOF) mass spectrometry (SELDI)‐MS ProteinChip technology. Sensitivity and specificity of each marker were calculated with receiver operating characteristic (ROC) curve methodology.
Results: Samples were divided into discovery and validation protein profile analysis studies. In discovery studies, we identified two protein species (6.7 kDa and 13.4 kDa) of significantly lower concentration in the CSF of ALS patients (n = 36) compared to normal control (n = 21) cases. Using the ROC analysis, we found that a combination of the 6.7 and the 13.4 kDa protein species (the ‘2‐protein’ model) predicted ALS more accurately than either protein species alone (92% accuracy, 91% sensitivity and 92% specificity in correctly identifying all ALS and non‐ALS patients). Independent validation studies using separate cohorts of ALS (n = 14), normal healthy controls (n = 25) and patients with pure motor and sensorimotor peripheral neuropathy (PN) (n = 7) confirmed the ability of the 6.7 kDa and 13.4 kDa CSF protein species to separate patients with ALS from other diseases.
Conclusions: This SELDI‐MS study suggests that additional application of ‘2‐protein’ biomarker model to current diagnostic criteria may provide an objective biomarker pattern to help identify patients with ALS.
Acknowledgement: This study was supported by MDA and Department of Veterans Affairs, ALS grant to GMP.
P97 THE SKELETAL MUSCLE TRANSCRIPTOME IN AMYOTROPHIC LATERAL SCLEROSIS
Gonzalez de Aguilar JL1, Halter B1, Dupuis L1, De Tapia M1, Di Scala F1, Wiederkehr C2, Demougin p2, Primig M2, Pradat PF3, Meininger V3 & Loeffler JP1
1INSERM U‐692, Université Louis Pasteur, Strasbourg, France, 2Pharmazentrum and Swiss Institute of BioInformatics, Basel, Switzerland, and 3Fédération de Neurologie Mazarin, Hôpital de la Pitié‐Salpêtrière, Paris, France
E‐mail address for correspondence: [email protected]
Background: The mechanism underlying ALS pathogenesis is still unclear but growing evidence suggests that initial alterations in skeletal muscle, preceding the onset of disease symptoms, may contribute to the disease process.
Objectives: We compared the muscle transcriptome of ALS‐related SOD1(G86R) mice with that of sciatic nerve‐axotomized mice. We also analyzed gene expression in skeletal muscle biopsies obtained from ALS patients and control subjects as a means to identify new potential markers of ALS severity.
Methods: We used a high‐density oligonucleotide microarray and applied advanced bio‐informatics protocols to provide a high throughput and accurate analysis of gene expression on a large scale.
Results: An unsupervised two‐dimensional clustering algorithm identified genes that are highly regulated in SOD1 (G86R) mice. While some of these genes represented the characteristic denervation process occurring in ALS, others were specifically regulated under the pathological condition but not after experimental denervation. In man, the statistical analysis revealed two groups of genes that categorized patients with defined characteristics. The first group included genes systematically repressed in all ALS patients, and the second one contained genes up‐regulated in correlation with the severity of motor impairment.
Conclusions: Skeletal muscle provides an additional source of information for the comprehension of the pathological mechanisms acting in ALS. The systematic approach at the gene level presented herein will serve to identify new targets for therapeutic intervention and diagnosis.
P98 PHASE 2B RANDOMIZED DOSE‐RANGING CLINICAL TRIAL OF TAMOXIFEN, A SELECTIVE ESTROGEN RECEPTOR MODULATOR [SERM], IN AMYOTROPHIC LATERAL SCLEROSIS (ALS): SENSITIVITY ANALYSES OF DISCORDANCE BETWEEN SURVIVAL AND FUNCTIONAL OUTCOMES WITH LONG‐TERM FOLLOW‐UP
Brooks B, Sanjak M, Roelke K, Parnell J, Peper S, Houdek A, Lunney D, Frye A, Robins I & Waclawik A
University of Wisconsin, Madison, USA
E‐mail address for correspondence: [email protected]
Background: ALS patients were assigned to blocks (n = 5) matched by age, sex, ALS onset site and baseline ALS Functional Rating Scale‐total score (ALS‐FRSt). Within each block, patients were randomized to one of five tamoxifen doses. Planned as a 12‐month clinical trial, it was extended to 24 months when there was no statistically significant change in endpoints (rate of computerized isometric muscle strength (CIMS) loss, vital capacity (VC) loss, ALS‐FRSt decrease) across the tamoxifen dose cohorts at 12 months.
Methods: Evaluations were at three‐ month intervals over 24 months. Follow‐up provided survival and event history data for achieving milestones in CIMS arm and CIMS leg, VC and ALS‐FRSt. Endpoints were analyzed by intention‐to‐treat allocation employing the Kaplan‐Meier method with log‐rank test and Cox proportional hazards modelling.
Results: ALS patients entered (34 males 26 females) had a mean age of 51 years, and ALS‐FRSt of 24 with no significant difference in these parameters among the five dose groups at baseline. Protocol analysis of the difference in survival for the two lower dose cohorts together versus the three upper cohorts together was significant (p<0.04 two‐sided safety p value). Cox proportional hazard analyses supported a drug dose effect. Prolongation of 80% survival benefit was above 200 days in the 20 mg, 30 mg, and 40 mg daily tamoxifen treatment cohorts. Survival in the 10 mg weekly‐10 mg daily cohorts was identical to that recorded for the ALS patients randomized to placebo in the Dutch ALS Creatine Study (1). No significant dose‐related adverse events occurred throughout the clinical trial. Safety analysis in a single‐site phase 2B dose‐ranging clinical trial of tamoxifen in ALS patients indicates a survival benefit of tamoxifen treatment with 20–30–40 mg daily compared with 10 mg weekly–10 mg daily. Sensitivity analyses do not support baseline age, sex, site of onset, duration of ALS, baseline ALS‐FRSt, baseline VC, or co‐morbidities as explanatory for this observed survival effect. Slower decrease in VC of the 20 mg, 30 mg, and 40 mg daily tamoxifen treatment cohorts in the second 12–24 month epoch might explain the survival differential but this was not statistically significant.
Conclusions: This survival benefit occurred in riluzole‐treated ALS patients and was as large or larger than the improved survival in riluzole‐treated compared with placebo‐treated ALS patients. This observation needs to be replicated in an appropriately statistically powered clinical trial to establish whether tamoxifen should be considered as an adjuvant therapy with riluzole in the treatment of ALS.
Acknowledgement: The study was supported by the Muscular Dystrophy Assocation, ALS Division, and National Institute of General Medical Sciences General Clinical Research Center. Tamoxifen was provided by Astra‐Zeneca Pharmaceuticals.
Reference
- Ann Neurol. 2003;53:437–45.
P99 A HISTORICAL DATABASE CONTROLLED PHASE II STUDY OF INTERFERON BETA 1A TREATMENT IN ALS
Simpson EP1, Yen AA2, Haverkamp L1 & Appel SH2
1Baylor College of Houston, Houston, and 2The Neurological Institute‐The Methodist Hospital, Houston, USA
E‐mail address for correspondence: [email protected]
Background: In ALS, the presence of innate and adaptive immune reactivity supports a potential role for pro‐inflammatory mechanisms in motor neuron injury. Interferon beta 1a (IFNβ1a), is an approved immuno‐modulatory drug for multiple sclerosis. Although the mechanisms of action are not completely defined, IFNβ1a clearly modulates both innate and adaptive immune responses, and may have beneficial effects in the treatment of ALS
Objective: To investigate the effect of IFNβ1a upon disease progression in ALS in a Phase II contemporary database‐controlled study.
Methods: A 12‐month Phase II single‐center, historical database controlled study of 60µg IFNβ1a and disease progression in sporadic ALS was used. Historical controls were derived from a contemporary database of the MDA/ALS clinic (Houston, TX), based upon study criteria. The primary outcome was post‐baseline slope of total Appel ALS score (AALS) to end of treatment. Secondary outcome measures included post‐baseline slope of AALS total score at 3, 6, 9, and 12 months; post‐baseline slope of the FVC, time to 20 point progression of AALS score, time to a FVC 50%, and survival. Statistics were based on an intent‐to‐treat principle using Student's t‐test or Mann‐Whitney test for group comparisons. Log rank analysis was used to compare times to clinical endpoints.
Results: Twenty‐nine patients were enrolled from July 2001 to July 2003. A contemporary, matched, database control group (n = 43) was selected according to inclusion/exclusion criteria. Fourteen patients completed the treatment period, with 21.4% discontinuing drug due to adverse events. The INFβ1a treated group showed a slower disease progression as measured by rate of change of AALS total score over 12 months compared to the database control population (3.76±3.56 pts/months vs. 4.94±3.42 pts/month; p = 0.042). Slower progression was also observed at three and six months with a trend toward significance at 9 months in the IFNβ1a treated group. No significant difference was observed in the time to 20 point progression in AALS total score from baseline between groups (p = 0.3), although the IFNβ1a treated group did show a delay in time to 20 point drop in FVC (logrank p = 0.027) and to 50% FVC from baseline (logrank p = 0.018) compared to the database population. Prolonged survival was associated with IFNβ1a therapy independent of pre‐slope and baseline AALS which were also found to be covariates of survival.
Conclusions: IFNβ1a treatment is associated with slowing of disease progression and prolongation of survival in an ALS patient population matched to a contemporary database‐controlled population. Thus, IFNβ1a appears to warrant further study as a potential therapy in ALS with a Phase III randomized placebo‐controlled trial.
P100 MULTI‐DRUG ‘COCKTAIL’ THERAPY IN ALS: A FEASIBILITY AND SAFETY STUDY
King RM & Rosenfeld J
Carolinas Medical Center, Charlotte, USA
E‐mail address for correspondence: [email protected]
Background: The use of multiple medications, supplements, and homeopathic compounds in the treatment of ALS is almost universal. Often patients and physicians devise their own cocktail of agents which are untested in controlled trials. The design and conduct of such trials are challenging and the feasibility of testing multi‐drug therapy needs to be better defined prior to onset of a clinical trial accessing efficacy.
Objectives: To assess the safety and feasibility of administering an eight‐drug combination trial in patients with ALS. Compliance was tested with two methods of drug administration. Safety issues (adverse events) were evaluated with a flow chart of anticipated side‐effects and a predetermined protocol of dose manipulation.
Methods: Eleven patients were enrolled in the open label multi‐drug study. Inclusion criteria included probable or definite ALS, forced vital capacity greater than 50%, disease duration of less than 5 years, and at least 5 or 10 upper extremity muscle groups with MRC grade of 4 or better. Patients were given medication instructions, photographs of the medication regimen for each time point (morning, lunch, dinner, evening), and diaries. Five patients received their medications pre‐dispensed in weekly medication dispensers and six patients received the medications in individual pill bottles. Patients returned at months 1, 2, 3, 5, 7, and 9 for follow‐up evaluation. Potential side‐effects were identified at the onset based on our experience with the components of the cocktail. Step‐wise procedures were established for dose reductions or cocktail modifications to accommodate adverse events.
Results: Patients were able to maintain the medication regimen and tolerated the large number of medications well. Compliance was greater in the group of patients that received the medication pre‐dispensed in weekly medication dispensers (99% compliance vs. 89% compliance). Four patients in the second group (pill bottles) began using pill dispenser or similar methods on their own. Patients were encouraged to keep a medication diary, recording any missed doses as well as any adverse events that occurred. Compliance with the medication diary was 75%. Our method of dose modification was successfully implemented for most adverse events. Individual components of the cocktail were, systematically, either reduced or stopped with appropriate resolution of the adverse event.
Conclusion: Patients were able to maintain compliance with the large number of medications and daily regimen, though providing patients with daily medication dispensers did improve compliance. Overall, adverse events occurred less often and with less severity than expected. The method of responding to adverse events with pre‐determined modification to the cocktail was effective and easily administered. The feasibility of designing and implementing a multi‐drug therapy is supported by these data. Subsequent study in an efficacy trial is warranted.
P101 FINDING AN INDIVIDUALIZED METHOD FOR MEASURING PROGRESSION IN ALS CLINICAL TRIALS
Schoenfeld DA1, Zhang H1, Kasarskis EJ2, Tandan R3, Cudkowicz ME1 & Shefner J3
1Massachusetts General Hospital, Boston, Massachusetts, 2University of Kentucky, Lexington KY, 3University of Vermont, Burlington VT, and 4Upstate Medical University, Syracuse NY, USA
E‐mail address for correspondence: [email protected]
Background: Clinical trials of ALS therapies often compare the slopes of an outcome measure between treatment groups to measure treatment efficacy. The variability of ALS presentations may cause some muscles to deteriorate initially more rapidly than others, and these may be predictable from observations before the trial begins.
Objectives: To generate an algorithm that finds a combination of sub‐measures that would change most rapidly for each subject, thus potentially increasing the power or reducing the sample size required for a treatment comparison.
Methods: Data were available on 488 patients from four clinical trials conducted by New England ALS consortium. Depending on the trial, measures were available either monthly or bi‐monthly. Three methods to develop an individualized score were attempted. We considered measurements used to screen patients and those made during the first two months of each clinical trial as if they were from a run‐in period, and subsequent measurements as if they were from the treatment period. The first method used the sum of the scores for muscles with the largest drop in value during the run‐in period as the individualized measure. The second used Lagrange multipliers to find the weighted sum of the sub‐measures that dropped the fastest during the treatment period subject to the constraint that the standard deviation of this weighted sum was constant, and that the weights were linear combinations of the measurements made during the run‐in period. The third used weights based on baseline values rather than run‐in values. We evaluated the methods using Lagrange multipliers by splitting the data into two subsets, a training sample used for determining the coefficients and a validation sample used for evaluating the method.
Results: The first method did not produce an advantage. The second method gave gains when applied to the training sample but not the validation sample. The third method using Lagrange multipliers applied to the baseline values resulted in a measure that reduced the sample size required to determine an effect of predetermined magnitude by 23% when applied to the validation sample.
Conclusions: Modest but important reductions in sample size are possible using this strategy. Using summary measures offers the prospect of combining qualitatively different measures (such as strength and score on a rating question) into a single measure that declines rapidly for an individual patient. The discrepancy found between the application of a measure to the training and validation set implies that a very large database is needed to train this algorithm. The estimated coefficients may provide useful information as to the way the disease symptoms progress.
P102 BULBAR FEATURE ASSESSMENT: A MULTILEVEL APPROACH
Ball LJ1, Green J2, Beukelman DR2 & Pattee GL1
1University of Nebraska Medical Center, Omaha, and 2University of Nebraska, Lincoln, USA
E‐mail address for correspondence: [email protected]
Background: People with bulbar features of ALS are likely to experience a mixed spastic‐flaccid dysarthria, a speech condition characterized by reduced intelligibility, at some point during disease progression. Existing staging systems are descriptive of current levels of speech functioning in people with ALS (PALS); however, they are not sensitive to early symptom detection. The bulbar neurons are prominent in differential diagnosis; therefore early identification of bulbar symptoms is crucial to facilitating diagnosis and increasing the earliest symptomatic PALS participation in pharmaceutical trials. In addition, early speech symptom indicators may assist in the development of effective treatments designed to sustain functional intelligibility.
Objectives:
Establish an assessment of bulbar features through a longitudinal study of clinical evaluations involving more than 300 PALS.
Identify procedures sensitive to identification of very early bulbar symptoms.
Demonstrate use of bulbar assessment for clinical trials.
Methods: In this session, a comprehensive assessment of bulbar attributes, comprised of selected levels of evaluation specificity, is tested: 1) Level 1, easily completed by clinical staff, is an assessment of speaking rate in sentences; B) Level 2, involves increased specificity in evaluation, through perceptual/acoustic measures of speech by a speech‐language pathologist; 3) Level 3, involves motion analysis of the tongue during speech using 3D articulography; 4) Finally, the three levels of assessment are implemented as part of bulbar function monitoring during a clinical trial.
Results: Data indicate that speaking rate is predictive of loss of speech intelligibility in PALS. Specific bulbar features include upper motor neuron and/or lower motor neuron characteristics observed in oral‐facial movements during functional activities (speech and swallowing). Data supporting the use of speaking rate as a measure of neurodeterioration in PALS are presented. A method was developed to quantify the diversity of tongue movement patterns for PALS spoken utterances. These findings are now being compared with reference data to gauge the degree of impairment of tongue function in speakers with neurologically‐based speech impairments, e.g. ALS. Results from computer based acoustic analysis indicate PALS demonstrates increased pause time and speech time during sentence production. Strategies to implement the bulbar assessment in a clinical trial, including frequency and appropriate timing of data collection, are presented.
Discussion: This project involves an effort to standardize assessment of bulbar function. Assessment and intervention strategies are included. Audio and video samples will be used for illustration of acoustic, perceptual and movement characteristics. Preliminary findings of this novel multilevel bulbar function assessment will be compared with the most common existing measures as applied during a clinical trial.
P103 STUDY OF RILUZOLE IN A PROSPECTIVE COHORT OF 501 SUBJECTS WITH AMYOTROPHIC LATERAL SCLEROSIS (ALS)
Qureshi MM1, Zhang H2, Shefner JM3, Schoenfeld DA2 & Cudkowicz ME1
1Department Of Neurology, Massachusetts General Hospital, Boston, 2Biostatistics Center, Department Of Medicine, Massachusetts General Hospital, Boston, and 3Department Of Neurology, SUNY Upstate Medical University, Syracuse, USA
E‐mail address for correspondence: [email protected]
Background: Riluzole prolongs survival in amyotrophic lateral sclerosis (ALS) as demonstrated in two clinical trials. An effect on functional measures was not shown in these two trials.
Objective: To examine the effect of riluzole on functional outcome measures in a database population of 501 ALS subjects enrolled in three recently completed clinical trials performed by the Northeast ALS Consortium (NEALS).
Methods: The randomized, double‐blind, placebo‐controlled clinical trials of topiramate, creatine and celebrex in ALS subjects were conducted from 1998 to 2004. Five hundred and one subjects diagnosed with probable or definite ALS were included in our analysis. We excluded the subjects randomized to topiramate because treatment may have accelerated disease course. Treatment duration was 12 months for participants in the topiramate and celebrex trials and six months for the creatine trial. Disease progression was measured with the Maximum Voluntary Isometric Contraction (MVIC) arm, grip megascores, the Amyotrophic Lateral Sclerosis Functional Rating Scale (ALSFRS) and vital capacity in all three trials; MVIC leg strength only in the celebrex trial. All subjects who took riluzole at baseline were considered as ‘riluzole users’.
A Kaplan‐Meier survival analysis based on riluzole treatment at the enrolment visit was performed. To control for baseline covariates, a Cox proportional hazards model was used. The rate of decline of different outcome measures was measured in slopes and compared in the riluzole and non‐riluzole groups. The difference was tested with t‐tests, and each subject's slope was estimated from the mixed model.
Results: Of the 501 ALS subjects, 318 took riluzole (riluzole group) whereas 183 did not receive riluzole at any stage of their enrolment period (non‐riluzole group). Survival increased approximately 19% in the riluzole group compared to the non‐riluzole group, but the difference was not significant (p = 0.32). The hazard function in the riluzole group did not change after controlling for various baseline covariates. There was no significant difference between the functional outcome measure slopes of the non‐riluzole and the riluzole groups. After controlling for the site of symptom onset, the MVIC arm megascore of subjects in the riluzole group (slope estimate = −0.091) declined more rapidly than the non‐riluzole group (slope estimate e = −0.079, p = 0.02).
Conclusions: In a cohort of 501 ALS subjects, riluzole was not found to influence rate of decline in muscle strength, pulmonary function or ALSFRS. A limitation of our study is that the three clinical trials were not randomized to riluzole use.
P104 HEPATITIS: AN ADVERSE EVENT DURING THE TREATMENT OF ALS WITH RILUZOLE OF HETEROGENEOUS ORIGIN
Pageot N, Barbaud A, Garrigues G & Camu W
MND Clinic, Department of Neurology, University Hospital Gui de Chauliac, Montpellier, France
E‐mail address for correspondence: [email protected]
Background: Riluzole has been licensed for ALS treatment since 1997. It is a well‐tolerated drug with a potential and rare liver toxicity, inducing cytolytic hepatitis which rapidly disappears after treatment withdrawal.
Objectives: To describe the spectrum of hepatitis during the treatment by riluzole in ALS.
Methods: In the last two years, nine patients (out of 300) were identified with a significant increase of either ALAT, ASAT, gamma glutamyl transferase (GGT) or alkaline phosphatase that led to them stopping riluzole or to decreasing their daily dosage.
Results: Seven males and 2 females aged from 42 to 78 years (mean 62 years) developed hepatitis. In one patient (case 1), sustained GGT was also associated with alcohol abuse. In three cases (cases 2, 3 and 4), ALAT and ASAT were increased above the 5N limit, corresponding well to previously described cases with cytolytic hepatitis. In five cases (cases 5 to 9), there was a significant increase of GGT (>5N except in 1 case>3N), without increase in ALAT or ASAT and the final diagnosis was cholestatic hepatitis. In cases 1, 2, 5 and 6 withdrawal of riluzole was not followed by complete normalization of the biological parameters as GGT after several months remained at 2N. In all the other cases biology is now normal.
Discussion: Our group of patients show that not only ALAT and ASAT should be screened during the course of ALS treatment by riluzole. While adverse biological events are not severe and do not appear to have any clinical consequence, the persistence of a GGT increase in four cases underlines the possibility that some ALS patients may have a liver disorder in parallel. Such liver abnormalities have been described in the literature but their origins remain undetermined. Recently the HFE gene has been described as a genetic susceptibility factor for ALS. It would be of interest to look for these mutations in our patients especially when the liver biology becomes abnormal.
Conclusion: Riluzole treatment remains well tolerated in ALS. Our cases underline the importance of a systematic check of GGT and alkaline phosphatase in parallel with ALAT and ASAT, which is not recommended to date by regulatory authorities. A liver disorder may evolve in parallel with ALS in some patients. However, we cannot suggest that such a disorder could play a role in ALS pathogenesis.
P105 THE RATE OF WEIGHT LOSS AND NOT BODY MASS INDEX IS A STRONG PROGNOSTIC FACTOR IN ALS
Chio A1, Calvo A1, Sellitti L1, Ghiglione P1, Terreni AA1, Galletti R2, Finocchiaro C2, Gamna F3 & Mutani R1
1Turin ALS Centre, Department of Neuroscience, University of Torino, Turin, 2Department of Clinical Nutrition, ASO San Giovanni Battista, Turin and 3Department of Respiratory Rehabilitation, ASO San Luigi Gonzaga, Orbassano (TO), Italy
E‐mail address for correspondence: [email protected]
Background: ALS prognosis is related to several factors, among which malnutrition seems to be one of the most relevant. However, there is no consensus about the best measure of malnutrition in ALS, which can be used in order to establish the necessity to start enteral nutrition via PEG/RIG.
Objective: To assess the indicators of malnutrition and outcome in a series of ALS patients followed up in our centre.
Methods: All patients regularly followed up in our centre, seen for the first time in the period 1 January 1996–30 June 2000, were included in the study. The patients were seen every three months and weight was evaluated at each visit. The patients also underwent ALS‐FRS. At the first visit, health body weight (HBW) was recorded. The following parameters (taken at diagnosis, at the six‐month and at the 12‐month visits) were considered for analysis: weight, BMI, weight loss or gain with respect to HBW, weight loss or gain with respect to the preceding visit. Possible confounding factors were considered, i.e. age, gender, type of onset (bulbar vs. spinal), presence of bulbar signs at the time of the visit, FVC% at the time of the first visit, PEG/RIG, non‐invasive ventilation, and tracheostomy. Survival analysis was performed by the Kaplan‐Meier method, and differences were assessed with the log‐rank test. Cox's multivariate analysis, stepwise, was performed to determine the independent prognostic factors.
Results: A total of 189 patients were included in the study (99 males and 90 females; mean age 61.5 years (SD 11.1), 63 bulbar onsets and 126 spinal onsets). The mean BMI when healthy was 26.5 (SD 3.9). The mean BMIs at diagnosis, at the six‐month, and at the 12‐month visits were 25.4 (SD 3.5), 24.6 (SD 3.8), and 24.0 (SD 3.7), respectively. At the six‐month and 12‐month visits, 13 and 19 patients, respectively, had a BMI lower than 18.5. The mean weight loss at the six‐month and 12‐month visit was 7.4% and 9.5%, respectively. The median survival from onset was 963 days. In univariate analysis, factors significantly related to outcome were age, site of onset, and FVC%, weight loss over 6% at the 12‐months visit (p<0.001) and rate of weight loss (i.e. mean weight loss per month) from onset to the six‐month and 12‐month visits. In multivariate analysis, the independent factors were mean weight loss per month (p<0.0001), FVC% (p = 0.0013), enteral nutrition (p = 0.003), and age (p = 0.017).
Conclusions: According to our data, the best indicator of malnutrition in ALS is the rate of loss of weight during the course of the disease. BMI did not result in a significant parameter for establishing the necessity of performing PEG, since several patients had a BMI over 18.5 when they had already lost more than 10% of their weight. Therefore, weight loss should be included in future trials as a reliable outcome parameter.
P106 SEVERE PROGRESSION OF AMYOTROPHIC LATERAL SCLEROSIS AFTER INTERVERTEBRAL DISCECTOMY
Stevic Z, Pavlovic S, Vujic A, Lavrnic D & Apostolski S
Clinical Centre of Serbia, Institute of Neurology, Belgrade, Yugoslavia
E‐mail address for correspondence: [email protected]
Background: The diagnosis of ALS can be mimicked in the initial phase by some other disorders such as intervertebral disc prolapse in the cervical spine with compressive radiculopathy/myelopathy.
Objectives: The aim of this study was to determine the frequency of ALS patients with intervertebral disc prolapse in the cervical spine with compressive radiculopathy and/or myelopathy and to determine specifities in the course of ALS in these patients.
Methods: The study population comprised 600 ALS patients diagnosed in the period 1992–2002 at The Institute of Neurology, Belgrade. All the patients fulfilled the El Escorial criteria of probable or definite ALS.
Results: Prolapse of intervertebral disc in the cervical spine with compressive radiculopathy and/or myelopathy was diagnosed in 47 (7.8%) ALS patients. Among them were 27 patients with severe pain in cervical spine and persistent paraesthesia in dermatomes C4/C5, C5/C6 or C6/C7, as initial symptoms. Intervertebral disc prolapse in the cervical spine with compressive radiculopathy and/or myelopathy was diagnosed within a period of 6–12 months, prior to the diagnosis of ALS. Wasting of the arm muscles was variable at the time of onset. After discus hernia with compressive radiculopathy and/or myelopathy had been diagnosed, seven patients underwent surgical treatment while 20 were conservatively treated. The mean age in the first subgroup was 49.1±10.4 years, and in the second subgroup 51.3±9.8 years (p>0.05). Patients from both subgroups developed progressive wasting in the arms and later of the limbs as well as bulbar muscles, and diagnosis of ALS was confirmed. Life span after the first symptoms had appeared in ALS patients who underwent discectomy was 21.1±1.6months. In ALS patients who had not been surgically treated it was 28.5±3.0 months. Comparison of these two subgroups of ALS patients showed significantly shorter lifespan in patients who underwent discectomy (p<0.01).
Conclusion: The results of our study showed that discectomy in the cervical spine seems to precipitate rapid progression of ALS.
P107 MILD EVOLUTION IN A SUBSET OF ALS/MND CASES WITH BULBAR ONSET: A DIFFERENTIAL DIAGNOSIS?
Pageot N, Barbaud A, Garrigues G & Camu W
MND Clinic, Department of Neurology, University Hospital Gui de Chauliac, Montpellier, France
E‐mail address for correspondence: [email protected]
Background: ALS is a severe disorder with, in approximately 25 to 30% of cases, a bulbar onset. In the latter presentation, the prognosis is the worst. In recent years, some cases of ALS/MND with a favourable evolution initially masquerading as ALS have been described.
Objectives: To describe a series of ALS/MND caseswith bulbar onset and mild evolution after one to six years.
Methods: In the last six years, eight patients were identified with a bulbar onset form of MND for whom a diagnosis of possible, probable or definite ALS was made. They were followed according to the requirements for ALS patients in our MND clinic.
Results: Three males and 5 females aged from 32 to 72 years (mean 55 years) were studied. Diagnosis of definite ALS was made in two cases, probable in two cases and possible in four cases. Symptoms at onset were dysphonia (three cases), dysphagia (three cases) and a mix of bulbar symptoms (two cases). In three cases EMG showed a denervation restricted to the bulbar level. In one case the disease was a pure lower MND. In six cases, upper motor neuron signs were present at three levels (bulbar, upper and lower limbs). Three patients clearly improved one year after diagnosis. One patient is stable after five years. For the four other cases, ALSFRS is declining at a maximum rate of 1 point per year. None of those cases are familial. Differential diagnosis used comprehensive biology, brain MRI, and lumbar puncture. For two males, a detection of the Kennedy's CAG expansion was performed and was negative. The third male had clear spasticity and this test was not proposed.
Discussion: While ALS is known to have a severe progression, slow ALS cases may represent 10 to 20% of the cases. The evolution of paralysis is a part of the diagnosis criteria, making it likely that a diagnosis of ALS may be refuted for a patient with mild evolution. We cannot propose, to date, another diagnosis than bulbar MND of slow or atypical ALS for these patients.
If one considers that the diagnosis ‘ALS’ cannot apply to those patients, this would imply that the diagnosis of ALS can only be an ‘a posteriori’ one. Even if this is the way of proceeding for the majority of neurologists, and if this is possibly the right way to proceed, an ‘a posteriori’ definition for a given concept is not scientifically correct.
Conclusion: Our eight bulbar onset ALS/MND cases either improved, did not worsen, or slowly evolved. It is tempting to propose an expansion of the concept of MND. Undoubtedly, this leads to being perhaps more careful than before when announcing a diagnosis or a prognosis to our patients.
P108 THE CLINICAL SPECTRUM OF SPORADIC ADULT‐ONSET UPPER MOTOR NEURON DISEASE
Brugman F1, Wokke JHJ1, Franssen H1, De Visser M2, De Jong JMBV2, Faber CG3, Kremer BP4, Schelhaas JH4, Bruyn RPM5, Kuks JBM6, Robberecht W7 & van den Berg LH1
1Rudolf Magnus Institute of Neuroscience, University Medical Centre Utrecht, Utrecht, 2Academic Medical Centre, Amsterdam, 3University Hospital Maastricht, Maastricht, 4University Medical Centre Nijmegen, Nijmegen, 5Diakonessenhuis Utrecht, Utrecht, 6University Medical Centre Groningen, Groningen, The Netherlands, and 7University Hospitals Leuven, Leuven, Belgium
E‐mail address for correspondence: [email protected]
Background: A syndrome of isolated upper motor neuron (UMN) degeneration was first described by Charcot in 1865 and was named primary lateral sclerosis (PLS) by Erb in 1875. It has been much debated whether PLS exists as a separate disease entity. Diagnostic criteria for PLS were proposed by Pringle et al. in 1992 and require progressive symmetrical spinobulbar spasticity of at least three years duration with adult onset, no clinical and at most minor electrophysiological evidence of lower motor neuron (LMN) involvement, a negative family history and exclusion of other causes.
Objectives: To explore the full clinical spectrum of sporadic adult‐onset UMN disease.
Methods: We performed a nationwide search for patients with adult‐onset sporadic UMN disease in the Netherlands. Inclusion criteria were: a gradually progressive UMN syndrome, adult onset, duration of three years or more, exclusion of ALS with needle EMG after at least three years duration, and absence of a family history suggestive of hereditary spastic paraparesis (HSP). Exclusion criteria were: clinical evidence of generalized LMN involvement, presence of cerebella, extrapyramidal, or sensory signs and evidence for other causes.
Results: Ninety‐seven patients were included (median age 58 years, range 22–80 years): 82 had onset in the lower limbs, four in the upper limbs, eight in the bulbar region and three simultaneously in different regions. At inclusion, 36 patients had bulbar involvement, 48 had involvement of upper limbs and 93 of lower limbs. Patients differed in the number of affected body regions, the chronological order of involvement of the different regions and the degree of symmetry. In some patients UMN signs were strictly unilateral or restricted to one region (lower limbs or bulbar). A transient phase of isolated lower limb involvement could be as long as 18 years. Minor EMG abnormalities were found in 40 of 78 re‐examined patients.
Conclusions: Patients with sporadic adult‐onset UMN disease show a wide spectrum of clinical presentations, with variable rates and patterns of progression. The Pringle criteria for PLS may need revision to include variants of UMN disease other than symmetrical spinobulbar spasticity. Clinical distinction with sporadic HSP remains problematic as long as DNA tests for most forms of HSP are not available. The presence of mild LMN signs in some patients is of uncertain significance, but could support the view that PLS is a clinical variant of ALS.
P109 PRIMARY LATERAL SCLEROSIS AND AMYOTROPHIC LATERAL SCLEROSIS DISPLAY DIFFERENCES AT DISEASE ONSET AND DURING FOLLOW‐UP THAT CAN HELP DIFFERENTIATE THE TWO ILLNESSES
Tartaglia MC, Rowe A, Findlater K & Strong MJ
Clinical Neurological Sciences, London Health Sciences Centre, London, Canada
E‐mail address for correspondence: [email protected]
Background: Primary lateral sclerosis (PLS) is a neurodegenerative disorder involving the upper motor system. Its existence as a separate entity from ALS is still controversial.
Objectives: To compare patients with PLS or ALS for differences in symptoms or signs at disease onset and during follow‐up to determine which features best differentiate these two entities.
Methods: Patients with PLS or ALS followed in the MND clinic were included for analysis. Mann‐Whitney tests were used to look for significant differences between the two groups.
Results: Seven hundred and thirty‐six patients with ALS and 47 patients with PLS were used for this analysis. The male to female ratio was not significantly different between the two groups. ALS patients were significantly older than PLS patients at symptom onset (59.17 vs. 54.43 years, p<0.005). Bulbar onset was more common in ALS patients in that 24% had dysarthria and 8.2% had dysphagia at symptom onset compared to 14.9% of PLS patients who had dysarthria and 0% presented with dysphagia (p<0.05). At presentation, paresthesias were not significantly different between the two groups (p = 0.37). Fasciculations and cramps were only a presenting feature in ALS patients with 8.8%, p<0.05 and 6.3%, p<0.05, respectively. Weakness was significantly more common as a presenting symptom in ALS patients at 56.7 vs. 42.6% (p = 0.05). There was a trend towards wasting being more common as a presenting sign in ALS patients at 6.1% (p = 0.053). Stiffness was significantly more common as a presenting symptom in PLS patients vs. ALS (46.8% vs. 3.3%, respectively, p<0.001). With respect to follow‐up, bulbar symptoms were significantly more common in ALS patients than PLS, at 87.5% vs. 74.5% (p<0.005). Limb wasting was rare in PLS patients, at 4.3% vs. 99.2% in ALS patients, p<0.001. The presence of pyramidal and sensory symptoms, however, was not significantly different between the two groups. Clinically overt dementia was significantly more prevalent in ALS patients vs. PLS, at 6.5% vs. 0%, p<0.05. Trauma was more commonly encountered in PLS patients, at 17% vs. 7.3 %, p<0.05. There was a trend for rheumatological disease to be more common in ALS patients, at 5% vs. 0% (p = 0.096). Autoimmune disorders were more common in PLS patients, at 4.3% vs. 0.4% in ALS patients (p<0.002).
Conclusions: PLS patients differ from ALS patients in their symptom onset site as well as their ongoing symptoms. Stiffness is more often a presenting symptom of PLS while fasciculations, cramps and wasting are uncommon in PLS. Over time, PLS patients rarely develop limb wasting compared with ALS patients.
P110 ASSESSMENT OF RESPIRATORY FUNCTION, SLEEP QUALITY AND ACTIVITIES OF DAILY LIVING IN PLS PATIENTS
Dellefave L, Wolfe L, Martinez J, Siddique N & Siddique T
Northwestern University Feinberg School of Medicine, Chicago, USA
E‐mail address for correspondence: [email protected]
Background: Primary lateral sclerosis (PLS) is a rare neurodegenerative disorder involving upper motor neuron (UMN) degeneration with slow symptom progression of lower extremities or bulbar region. A paucity of respiratory function and sleep data is reported for PLS patients. With characterization of PLS respiratory involvement, discrete differences in respiratory function between UMN and lower motor neuron involvement may be identified.
Objective: To assess the impact of PLS on sleep and pulmonary function using spirometry, St. George Respiratory Questionnaire (SGRQ), Pittsburgh Sleep Quality Index (PSQI), and ALS Functional Rating Scale‐Revised (ALSFRS‐R), compared with ALS control subjects.
Methods: Recruitment of 11 PLS and 11 ALS patients was through PLS Patient Conference and MND clinics at one site in the USA. Respiratory function was measured as forced vital capacity (FVC). ALS controls matched FVCs of PLS subjects. Subjects self‐administered three questionnaires: SGRQ measures respiratory involvement, PSQI measures sleep quality and ALSFRS‐R. Validated electronic databases scored SGRQ PSQI and ALSFRS‐R. Student's two‐tailed t‐tests analyzed data of PLS vs. ALS subjects.
Results: PLS subject FVCs are not significantly different from those of ALS controls (mean 93.45 and 79.45, respectively, t = 1.81, p = 0.086). Duration of disease is not significantly different between PLS and ALS (mean 87.4 and 54.1 respectively, t = 1.53, p = 0.07) in our study population. ALS subjects had significantly higher SGRQ symptoms score (t = 2.47, p = 0.022). ALS subjects had a significantly higher total SGRQ score (t = 2.12, p = 0.046). SGRQ activity and impact components (demonstrating quality of life) are not significantly different in PLS compared with controls (t = 1.67, p = 0.11; t = 1.57, p = 0.13, respectively).
Both groups had PSQI total values above 5, signifying ‘poor sleep’. None of the PSQI measures of sleep quality are significantly different for PLS compared to ALS; however, the PLS group had shorter total sleep time than ALS (mean 0.73, 1.27 respectively, t = 1.27, p = 0.22), reduced sleep efficiency (mean 0.56, 0.91 respectively, t = 0.86, p = 0.40), and used more sleep medications (mean 0.64, 0.27 respectively, t = 0.88, p = 0.39). None of the ALSFRS‐R measures were significantly different in PLS vs. ALS.
Discussion: ALS and PLS subjects are well matched by FVC to eliminate disease severity as a cause of differences between the two disease groups. PLS subjects were symptomatic longer than equivalent ALS, suggesting respiratory impairment is seen much earlier and is more severe compared to PLS. PLS subjects had fewer respiratory complaints then ALS, suggesting there may be unmeasured factors playing a role in neuromuscular control of respiration not estimated by FVC measurement alone. Based on publications, FVC can underestimate the degree of respiratory impairment in ALS. Of note, PLS subjects reported a reduction in overall sleep quality compared to ALS, even though they used more sleep promoting medications.
P111 FLAIL‐LIMB VARIANT OF MOTOR NEURON DISEASE HAS EVIDENCE OF CENTRAL AND PERIPHERAL NERVE DYSFUNCTION
Vucic S & Kiernan MC
Prince of Wales Medical Research Institute, Institute of Neurological Sciences, University of NSW, Sydney, Australia
E‐mail address for correspondence: [email protected]
Background: Motor neuron disease (MND) is a progressive neurodegenerative disorder of corticomotoneurons and anterior horn cells (AHC), with median survival of approximately two years. In 20% of MND patients, survival is prolonged greater than five years. A specific group of MND patients with increased survival is the flail limb variant of MND, also known as ‘man‐in‐the barrel’ syndrome or brachial amyotrophic diplegia syndrome (BAD). This variant of MND is characterized by proximal onset of muscle weakness and wasting in the upper limbs with initial preservation of bulbar and lower limb function. The aim of the present study was to further define the clinical phenotype of the flail limb variant of MND, and to explore disease pathophysiology through the application of novel nerve excitability and transcranial magnetic threshold tracking techniques.
Methods: Clinical data, structural imaging and conventional neurophysiological studies were collected for patients with the flail limb variant of MND. Inclusion criteria included a pattern of weakness confined to the proximal aspects of the upper limbs for at least two years. The results of neurophysiological testing were used to exclude patients with multifocal motor neuropathy with conduction block and patients with other forms of demyelinating neuropathy. Patients were followed with the amyotrophic lateral sclerosis functional rating scale‐revised (ALS‐FRS‐R). Nerve excitability testing was undertaken in a proportion of patients.
Results: In total, nine patients with the flail limb variant of MND were detected from our database. Their mean age at disease onset (60.3 years) was similar to patients with other variants of MND. The male: female ratio was significantly increased in the flail limb variant of MND (8:1 vs. 1.6:1, p<0.05). Mean duration of illness was also significantly longer in the flail limb variant of MND (76.3 vs. 20.8 months, p<0.05). Lower limb symptoms developed in two (29%) patients 4.9 years after symptom onset, while bulbar symptoms developed in three (43%) patients, 36 months after symptom onset. Respiratory symptoms occurred in two patients, 31 months after symptom onset and both patients required non‐invasive ventilatory support. Transcranial magnetic stimulation revealed absence of early intracortical inhibition consistent with cortical hyperexcitability in the absence of clinical upper motor neuron signs. Nerve excitability testing demonstrated evidence of activity dependent conduction failure, suggesting the presence of axonal membrane Na+/ K+ ATPase pump dysfunction.
Conclusion: The flail limb variant of MND is a distinct clinical type of MND with male predominance and longer survival. Cortical excitability is increased, similar to patients with other forms of MND, arguing against isolated lower motor neuron involvement.
P112 CLINICAL AND NEUROPATHOLOGICAL STUDY OF A SERIES OF ALS‐DEMENTIA SYNDROME PATIENTS: ANALYSIS OF OUR EXPERIENCE
Gamez 1, Ortega‐Aznar A2, Pujadas F1, Corbera M1, Jacas C3, Muñoz X4 & Alvarez‐Sabin J1
1ALS Unit, Neurology Department, 2Neuropathology Department, 3Neuropsychology Department, and 4Pneumology Department, Hospital Universitari Vall d'Hebron, Barcelona, Spain
E‐mail address for correspondence: [email protected]
Background: ALS has traditionally been considered a neurodegenerative disease that selectively affects the motor neurons. However, some patients show evidence of cognitive impairment, which in most cases is mild and appears in the advanced stages of the disease. Dementia is nevertheless severe in a subset of patients and appears at the same time as motor deficits, and is therefore an ALS‐dementia syndrome (ADS). We present the findings for 13 patients with ADS.
Objectives: Clinical and neuropathological characterization of the ALS‐dementia syndrome.
Patients and methods: We identified 13 individuals with clinical ADS criteria in the database of our ALS Unit (417 patients). They were neuropsychologically assessed. Screening for SOD1 gene mutations was carried out. A necropsy was performed on six patients.
Results: The prevalence of ADS was 3.1%. The average age of onset was 63.3 years and the average survival time was 27.9 months. Bulbar forms predominated. There was a fronto‐temporal dementia (FTD) profile in 11 patients. The atypical clinical findings observed were supranuclear gaze paralysis in one patient and voluntary lid closing inability (eyelid ‘apraxia’) in another. No mutations in SOD1 gene were found.
As regards neuropathological findings, we observed varying degrees of spongiosis in layers I/II of the frontal cortex, neuronal loss and presence of subcortical gliosis in all patients. We identified ubiquitin‐immunoreactive inclusions in the cortical neurons in five patients. In the spinal cord, neuronal loss and gliosis in the ventral horn was most severe in the cervical region in three patients, in the dorsal region in two, and the lumbar region in one. We found atrophy of the ventral roots and severe motor neuron loss in the hypoglossus nucleus in all patients. Three patients presented substantia nigra neuronal loss. Neuritic plaques in the hippocampus were identified in two patients.
Discussion: These findings suggest that ALS could be a multisystemic disorder with a limited phenotypical expression. The most striking clinical finding was the presence of ocular‐motor abnormalities in two patients of the series. The neuropathological abnormalities present in ADS would represent a ‘continuum’ with other neurodegenerative diseases.
Acknowledgement: FIS 02/0648
P113 MUSCLE AND JOINT PAIN IN MND/POST‐POLIOMYELITIS (PPS)
Quadros AAJ, Silva TM, Ramos PG, Pereira RDB, Silva HCA & Oliveira ASB
1Universidade Federal de São Paulo/Escola Paulista de Medicina, São Paulo, Brazil, and 2ABRASPP ‐ Post‐Polio Brasilian Association, São Paulo, Brazil
E‐mail address for correspondence: [email protected]
Introduction: Post‐polio syndrome (PPS) is the term most commonly used to describe the new difficulties which may occur many years after recovery from paralytic poliomyelitis. Although many symptoms may be reported by these patients, the most commonly reported are new weakness, fatigue, and pain. These three symptoms have been used to define PPS.
The cause of PPS remains unclear, although the most widely accepted hypothesis, proposed by Wiechers and Hubbell, attributes these symptoms to a distal degeneration of axons in the greatly enlarged motor units that developed during recovery following acute paralytic poliomyelitis.
Currently, it is unclear how the symptom of pain, which is uncommon in the classical motor unit disease such as amyotrophic lateral sclerosis, is related to disease of the motor unit. Although in ambulatorial assessment pain is frequently reported, pain can be due a variety of causes. It can be caused by muscular abnormalities, joint and soft tissues abnormalities, and other superimposed neurological abnormalities. Possible causes for muscular pain include muscular overuse, muscular cramps, fasciculation, and fibromyalgia.
Objectives: The purpose of this study is to demonstrate the pain features in PPS ambulatorial patients.
Patients and methods: Between March 2003 and June 2004, 129 patients (81 female, 48 male, age 18 to 72 years, mean 39.9±10.07 years), with a confirmed diagnosis of PPS according to consensus criteria, attended the outpatient ambulatory neuromuscular disease of the School of Medicine of University of São Paulo. All patients submitted to neurological evaluation, blood and urine tests and their pain was evaluated using the Pain Analogical Scale (PAS).
Results: Joint pain was found in 79.8% (103) and muscle pain in 76.0% (98) of patients. The PAS scores ranged from 0 to <10. A pain intensity score of 4 was found in 10.9%, and a score 5 in 14.7% (moderate pain). 14.7% had a pain score of 7 and 18.6% a score of 8, representing moderate to high pain. The distribution of pain found was: 8.5% (12) no pain; fibromyalgia 2.3% (3); headache 2.3% (3); cervical pain 2.3% (3); shoulder pain 7.0% (9); upper limb pain 19.4% (25).