THEME 5 IMAGING, ELECTROPHYSIOLOGY AND MARKERS OF DISEASE PROGRESSION

P58 PATHOPHYSIOLOGICAL INSIGHTS INTO FACIAL ONSET SENSORIMOTOR NEURONOPATHY SYNDROME: A NOVEL SYNDROME IN NEUROLOGY

VUCIC S¹, REDDEL S², TISCH S³, KOTSCHET K⁴, KIERNAN M⁵

¹Sydney Medical School Westmead, University of Sydney, Sydney, NSW, Australia, ²Concord General and Repatriation Hospital, Sydney, NSW, Australia, ³St Vincent's Hospital, Sydney, NSW, Australia, ⁴St Vincent's Hospital, Melbourne, Victoria, Australia, ⁵Neuroscience Reserach Australia, University of New South Wales, Sydney, NSW, Australia

Email address for correspondence: [email protected]

Keywords: FOSMN, cortical excitability, neurodegeneration

Background: Facial onset sensory and motor neuronopathy (FOSMN) syndrome is a slowly progressive neurodegenerative disorder of sensory and motor neurons, heralded by development of sensory deficits in the trigeminal nerve distribution, with spread of symptoms to affect the neck, upper trunk and upper limbs in sequential order. Motor deficits develop later in the course of the disease. Although degeneration of sensory and motor neurons underlie the development of FOSMN syndrome, the pathophysiological mechanisms remain to be determined. Recently, the co-existence of upper and lower motor neuron signs in FOSMN syndrome suggested a link between FOSMN syndrome and amyotrophic lateral sclerosis (ALS).

Objectives: Given that cortical hyperexcitability appears to be intrinsic to the pathophysiology of ALS, the present study aimed to assess the pathophysiological mechanisms underlying FOSMN syndrome, and in particular whether cortical hyperexcitability is a prominent feature in FOSMN syndrome.

Methods: Studies were undertaken on five patients with clinical features of FOSMN syndrome, heralded by onset of sensory deficits in the trigeminal nerve distribution. All patients underwent clinical, laboratory, conventional neurophysiological and neuroradiological assessment to exclude potential mimic disorder of FOSMN syndrome. Cortical excitability studies were undertaken using a threshold tacking transcranial magnetic stimulation (TMS) technique utilising a 90 mm circular coil. Recordings of motor evoked potential (MEP) responses were made over the abductor pollicis brevis. Results were compared to 30 age-matched healthy subjects and 104 ALS patients.

Results: There were three male and two female patients with FOSMN syndrome with mean disease duration of 52.3 ± 10.6 months. Upper motor neuron signs were not evident in any of the FOSMN patients. Conventional neurophysiological studies revealed abnormalities of blink reflexes, prolonged or absent R1 and R2 components, in all FOSMN patients tested. Pathological studies disclosed the presence of axonal degeneration with absence of inflammation or amyloid deposition. Cortical excitability studies revealed that the resting motor threshold was increased in FOSMN patients (71.8 ± 4.3%) when compared to ALS (57.2 ± 0.9%, P < 0.001) and controls (61.8 ± 1.6%, F = 8.7, P = 0.05). In addition, averaged short interval intracortical inhibition was increased in FOSMN patients (6.8 ± 1.3%) when compared to ALS (2.6 ± 0.9%, F = 11.2, P < 0.05) but comparable to healthy controls (10.5 ± 1.1%, P = 0.09). Of relevance, there was a significant difference between groups in the cortical silent period duration (FOSMN 204 ± 12.7 ms; ALS 181.3 ± 4.3 ms; controls 215.3 ± 3.8 ms, F = 8.5, P < 0.001), and MEP amplitude (FOSMN 26.6 ± 7.0%; ALS 38.1 ± 2.2%; controls 25.2 ± 2.8%, F = 4.7, P < 0.05).

Discussion and conclusions: The findings in the present study suggest that although FOSMN syndrome is a slowly progressive neurodegenerative disorder of sensory and motor neurons, cortical hyperexcitability does not appear to underlie the development of FOSMN syndrome. In addition, the threshold tracking TMS technique clearly differentiated FOSMN syndrome from ALS, thereby arguing against a potential link between these two disorders.

P59 DISSECTING CENTRAL AND PERIPHERAL MOTOR CONTRIBUTIONS TO THE PATHOPHYSIOLOGY OF SPINAL MUSCULAR ATROPHY

FARRAR M^1,2, VUCIC S¹, KIERNAN M¹

¹Neurosciences Research Australia and University of New South Wales, Randwick, NSW, Australia, ²Sydney Children's Hospital, Randwick, NSW, Australia

Email address for correspondence: [email protected]

Keywords: transcranial magnetic stimulation, spinal muscular atrophy

Background: Spinal muscular atrophy (SMA) clinically presents as a pure lower motor neuron disorder with muscle weakness and atrophy. Neurodegeneration of spinal motor neurons in SMA is secondary to reduced levels of the survival motor neuron (SMN) protein, which is also expressed in the brain. Whether cortical dysfunction or alternatively, plasticity occurs in response to spinal motor neuron neurodegeneration remains to be determined.

Objectives: To gain further insights into corticomotoneuronal function and disease pathogenesis in SMA, the present study utilised clinical and functional assessments, combined with threshold tracking transcranial magnetic stimulation (TMS) techniques and peripheral nerve studies to investigate SMA patients.

Methods: Cortical and peripheral nerve excitability studies were undertaken in 11 SMA patients with homozygous deletions in the SMN1 gene (mean age 21.9 years, range 16-36 years) and 24 age matched controls. A comparison was also made with a disease control group, comprising 81 amyotrophic lateral sclerosis (ALS) patients. Motor-evoked potentials and compound muscle action potentials (CMAPs) were recorded from the right abductor pollicis brevis. Neurophysiological parameters were correlated with clinical measures of disease severity.

Results: Motor evoked potential amplitude was significantly increased in SMA when compared to healthy controls, but similar to ALS (SMA 39.7 ± 4.0%; ALS 38.8 ± 2.8%; controls 20.3 ± 2.5%; F = 10.1, P < 0.0001). In contrast, short interval intracortical inhibition (SMA 14.4 ± 1.6%; ALS 4.3 ± 1.8%; controls 17.0 ± 2.3%, F = 11.4, P < 0.0001) and cortical silent period duration (SMA, 204.4 ± 9.8 ms; ALS, 182.7 ± 5.2 ms; controls 208.8 ± 3.7 ms, F = 4.8, P = 0.01) were similar between SMA patients and healthy controls, but significantly larger when compared to ALS. Of relevance, peripheral disease burden, as measured by the CMAP amplitude (SMA, 6.3 ± 0.8mV; ALS, 5.9 ± 0.4 mV; controls, 11.8 ± 0.5 mV, F = 35.5, P < 0.0005) and neurophysiological index (SMA, 0.7 ± 0.2; ALS, 0.7 ± 0.1; controls, 3.1 ± 0.2, F = 108.2, P < 0.0005), were significantly reduced in both SMA and ALS patients when compared to healthy controls.

Discussion and conclusions: Simultaneous assessment of central and peripheral motor pathways has established degeneration of the lower motor neuron axis. Despite this process, there remains preservation of corticomotorneurons in SMA, refuting any significant physiological effects of reduced SMN protein expression within the CNS. These findings suggest a unique pathophysiology for motor neuron degeneration in SMA, particularly when compared to ALS. Further, the descending corticomotorneuron output is preserved in SMA and consequently its greater contribution onto the surviving spinal motor neurons likely represents adaptive neuroplasticity.

P60 UPPER MOTOR NEURON ABNORMALITIES IN PARRY-ROMBERG SYNDROME

WINHAMMAR J^1,2, VUCIC S², KIERNAN M², ROWE D^1,3

¹Departments of Neurology, Royal North Shore Hospital, St Leonards, NSW, Australia, ²NEURA and Prince of Wales Clinical School, Randwick, NSW, Australia, ³Australian School of Advanced Medicine, Macquarie University, Australia

Email address for correspondence: [email protected]

Keywords: Parry Romberg Syndrome, transcranial magnetic stimulation, diffusion tensor imaging

Background: Parry-Romberg Syndrome (PRS) is a clinically heterogenous disorder characterised by hemifacial atrophy, and variable associated intracerebral abnormalities that produce contralateral neurological manifestations, including hemiparesis, hemianopia and focal seizures. The aetiology of PRS is unknown, but there are a few reported associations with scleroderma.

Objectives: A subject with PRS had assessment of the upper motor neurons (UMN) with diffusion tensor imaging (DTI) and transcranial magnetic stimulation (TMS) to identify abnormalities in the motor cortex and in the corticospinal tract. The data from this patient with PRS is also compared with normal subjects and subjects with UMN abnormalities as a part of Amyotrophic Lateral Sclerosis (ALS).

Methods: TMS was applied to the motor cortex by a 90mm circular coil connected to a BiStim device, with magnetic evoked potentials recorded from abductor pollicis brevis. Threshold tracking was performed using a paired-pulse paradigm with a conditioning stimulus delivered at intervals before a suprathreshold test stimulus. MRI studies were performed using a 3 Tesla Philips Intera system (Philips Medical Systems, Best, The Netherlands) with an eight channel, phased array head coil and gradient coils (0-33mT/m).

Results: There were differences between the right and left cerebral cortex in PRS as measured by TMS. The Resting Motor Threshold (RMT) was higher in the right hemisphere (90%) than in the left hemisphere (74%), while the Central Motor Conduction Time (CMCT) was prolonged in the right hemisphere (7.95 ms) than in the left hemisphere (3.7 ms). In addition, the Motor Evoked Potential amplitude in the right hemisphere (1.2 mV) left hemisphere (4.4 mV) and Short-Interval Intracortical Inhibition were lower in the right hemisphere (−5.58% threshold change) left hemisphere (4.25% threshold change).

MRI examination of the brain identified generalised atrophy of the right cerebral hemispheres, while the left cerebral cortex was anatomically normal. In addition, DTI studies disclosed a significant reduction of fractional anisotropy using fiber tracking (left cerebral peduncle = 0.64, right cerebral peduncle = 0.5) in the PR patient and (left cerebral peduncle = 0.64 and right cerebral peduncle = 0.60) in the control group and increase in perpendicular diffusivity in the PR patient (left hemisphere 0.49×10-3 mm2/s and 0.38×10⁻³ mm²/s in the right hemisphere) when compared to controls in the right hemisphere (left hemisphere 0.375×10⁻³ mm²/s and right hemisphere 0.41×10⁻³ mm²/s).

Discussion: The findings in the present study indicate anatomical and functional abnormalities of the right cerebral hemisphere contralateral to the side of deficit. Interestingly, cortical function was preserved in the left hemisphere, the side ipilateral to the limb deficits. The finding of cortical dysfunction is most consistent with developmental abnormalities, possibly occurring prior to decussation of pyramidal fibers in utero. Taken together, these findings seem to suggest that Parry-Romberg syndrome may be a developmental disorder occurring during organogenesis in some patients.

P61 CLINICAL AND NEUROPHYSIOLOGICAL EVIDENCE OF MOTOR NEURON DYSFUNCTION IN FRONTOTEMPORAL DEMENTIA

BURRELL J^1,2, KIERNAN M^1,2, VUCIC S^1,3, HODGES J¹

¹Neuroscience Research Australia, NSW, Australia, ²Prince of Wales Clinical School, Prince of Wales Hospital, NSW, Australia, ³Sydney Medical School, Westmead Hospital, NSW, Australia

Email address for correspondence: [email protected]

Keywords: biomarkers, frontotemporal dementia, transcranial magnetic stimulation

Background: Frontotemporal dementia (FTD) and motor neuron disease (MND) share clinical and pathological characteristics. MND develops in some patients with FTD, but the incidence, severity and functional significance of motor dysfunction in FTD has not been determined.

Objectives: To identify and characterise clinical and neurophysiological evidence of motor neuron dysfunction in FTD.

Methods: We performed detailed clinical and neurophysiological assessments on 40 consecutive FTD patients, 42 age/gender matched MND patients and 26 control subjects. The neurophysiological index (NI) was used to detect lower motor neuron dysfunction. Short interval intracortical inhibition (SICI), measured using paired pulse threshold tracking transcranial magnetic stimulation, was used to document upper motor neuron dysfunction.

Results: Of 40 FTD patients, 5 (12.5%) developed concomitant MND (FTD-MND) during the course of the study. A further 9 (27.3%) FTD patients showed clinical evidence of motor neuron dysfunction (ie wasting, fasciculations or weakness), which was generally mild. The neurophysiological index was reduced in FTD (1.1 +/− 0.9) compared to controls (1.9 +/− 0.8, P < 0.001), but relatively preserved compared to MND (0.7 +/− 0.6, P < 0.05). Average SICI was reduced in FTD (4.3 +/− 1.7%) compared to controls (9.1 +/− 1.1%, P < 0.05), and similar to MND: it was particularly reduced in patients with FTD-MND and progressive non-fluent aphasia, but was normal in other FTD variants.

Conclusions: In this cohort of 40 consecutive FTD patients, 5 (12.5%) developed FTD-MND and a further 9 (27.3%) had clinical evidence of mild motor neuron dysfunction. Neurophysiological evidence of lower and upper motor neuron dysfunction was common, even in FTD patients with no clinical evidence of MND. Upper motor neuron dysfunction was most marked in patients with progressive non-fluent aphasia.

P62 THE SPLIT HAND INDEX: A POTENTIAL DIAGNOSTIC TEST FOR ALS

MENON P¹, VUCIC S¹, YIANNIKAS C¹, STROUD J¹, KIERNAN M¹, CHEAH B¹

¹Westmead Hospital, Sydney, NSW, Australia, ²Southern Neurophysiology, Burwood, Sydney, NSW, Australia, ³Neuroscience Research Australia, Randwick, Sydney, NSW, Australia

Email address for correspondence: [email protected]

Keywords: novel, diagnostic, tool

Objective: Preferential atrophy of the Abductor Pollicis Brevis (APB) and the First Dorsal Interosseous (FDI) is a classical feature of amyotrophic lateral sclerosis (ALS), and termed the split-hand. The present study assessed the diagnostic utility of a novel neurophysiological measure, the split-hand index (SI), in ALS and in particular, whether SI could reliably differentiate ALS from mimic disorders.

Methods: 20 ALS and110 non-ALS patients were recruited into the study according to the Standards for Reporting of Diagnostic Accuracy (STARD) criteria. Baseline-peak compound muscle action potential (CMAP) amplitudes were recorded over the APB, FDI and Abductor Digit Minimi (ADM) muscles. The SI was calculated as follows:

SI = (CMAP_{APB AMPLITUDE} × CMAP_{FDI AMPLITUDE})/CMAP_{ADM AMPLITUDE}

Results: SI was significantly reduced in ALS patients compared to controls (ALS 3.2 ± 0.6; non-ALS 9.0 ± 0.5, P < 0.0001). Analysis of receiver operating characteristic curves suggested that SI reliably and robustly distinguished ALS from mimics (area under curve 0.90, P < 0.0001). The optimal SI value for differentiating ALS from non-ALS was 5.2, with a sensitivity of 82%, specificity of 81%, positive likelihood ratio 4.4 and negative likelihood ratio of 0.23. Of relevance, the SI correlated with established clinical (APB Medical Research Council score, R = 0.7) and neurophysiological biomarkers (R = 0.44).

Conclusion: Findings from the present study establish that the split hand index is a simple and robust diagnostic neurophysiological test clearly differentiating ALS from mimic disorders and may enable earlier diagnosis in ALS and therefore institution of neuroprotective therapies and recruitment into clinical trials.

P63 THE “SPLIT-HAND” PATTERN OF ALS: AN ELECTRODIAGNOSTIC SIGN OF ALS?

PIORO E¹, ZWICKER J²

¹Cleveland Clinic, Cleveland, OH, USA, ²Ottawa Hospital, Ottawa, ON, Canada

Email address for correspondence: [email protected]

Keywords: nerve conduction study, intrinsic hand muscles, ulnar neuropathy

Background: The “split-hand” pattern (SHP) is an electrodiagnostic (EDx) finding of predominant denervation in thenar and/or first dorsal interosseus (FDI) muscles relative to hypothenar muscles, with normal sensory responses. Although reported as highly suggestive of ALS/MND, an axon loss ulnar neuropathy causing more denervation in FDI than in hypothenar muscles could result in SHP.

Objective: Determine frequency of the SHP in ALS and in ulnar neuropathies.

Methods: Median and ulnar nerve EDx studies in Cleveland Clinic EMG laboratory during five-year period of ALS patients (definite or probable by El Escorial criteria) and of patients with ulnar neuropathy were reviewed. Compound muscle action potentials (CMAPs) were recorded from thenar muscles (stimulating median nerve at wrist), and from FDI and hypothenar muscles (stimulating ulnar nerve at wrist). Sensory nerve action potentials (SNAPs) were recorded from 2^nd and 5^th digits antidromically, stimulating median or ulnar nerve at wrist, respectively.

Results: 154 EDx examinations recording CMAPs from 252 hands of 142 ALS patients revealed: 1) low CMAPs in FDI, hypothenar and thenar muscles in 32% (81/252), 2) low thenar and FDI CMAPs with normal hypothenar CMAPs in 14% (35/252), 3) low FDI CMAPs alone in 9.5% (24/252), 4) low thenar CMAPs alone in 5.5% (14/252), 5) low hypothenar CMAPs alone in 0% (0/252), 6) normal CMAPs in all three muscle groups in 26% (66/252). Of 81 hands with diffusely low CMAPs but hypothenar CMAPs ≥ 2mV, the FDI CMAP was < 50% of hypothenar CMAP in 17 hands (21%). Ninety ulnar neuropathies (in 82 patients) revealed that CMAPs were normal in 17% (15/90), low in FDI in 11% (10/90), in ADM in 13%, and in both FDI and ADM in 59%. Ulnar SNAPs were normal in 24% (22/90); of these, CMAPs were normal in 27%, low in FDI in 18% (4/22), in ADM in 14%, and in both FDI and ADM in 36%. Of cases with only low FDI CMAP, lesion localization was at/distal to the wrist in 33% (3/10), at the forearm in 20%, at/proximal to the elbow in 20%, and not possible in 33%. Three of 4 cases with low FDI CMAPs alone and normal SNAPs were due to lesions in the hand.

Discussion: In ALS/MND patients, CMAPs were preferentially low in lateral compared to medial hand muscles in 36% of hands (90/252). In contrast, only 4% (4/90) of ulnar neuropathies met SHP definition with preserved ulnar SNAPs; all but one was from a lesion in the hand.

Conclusions: Approximately 1/3 of patients with definite or probable ALS have the SHP. It can occur in ulnar neuropathies but is only a “partial” SHP with sparing of thenar muscles. However, limited EDx abnormalities would differentiate these cases from ALS and further support the SHP as highly suggestive of ALS.

P64 THE DISTRIBUTION OF ELECTRODIAGNOSTIC ABNORMALITIES IN EARLY AMYOTROPHIC LATERAL SCLEROSIS

SIMON N¹, KIERNAN M¹

¹Prince of Wales Clinical School, University of New South Wales, Randwick, NSW, Australia, ²Neuroscience Research Australia, Randwick, NSW, Australia, ³Institute of Neurological Sciences, Prince of Wales Hospital, Randwick, NSW, Australia

Email address for correspondence: [email protected]

Keywords: diagnosis, electromyography, neurophysiology

Introduction: Since the Awaji-shima consensus criteria, detection of neurogenic change on needle electromyography has been added to the clinical El Escorial diagnostic criteria for amyotrophic lateral sclerosis (ALS) as evidence of lower motor neuron involvement. In patients with clinically suspected ALS, the relative weighting of upper and lower motor neuron involvement can make diagnosis difficult. Hence it is important that electromyography is focussed on the muscles of highest yield in order to maximise the sensitivity of detection of neurogenic change in segments where lower motor neuron involvement is not readily apparent clinically.

Methods: Electrodiagnostic and clinical data were collected prospectively from 142 patients with clinically suspected ALS in a specialised neurophysiology department attached to a tertiary referral ALS clinic. The involved segments at onset and relative distribution of clinical changes at the time of electrodiagnostic testing were recorded. A standard series of sensory and motor nerve conduction studies and electromyography sampling was performed.

Results: Electrophysiological data was obtained a mean of 14 months after symptom onset. Tibial nerve motor responses were abnormal in 25% of nerves (33% of patients) with reduced compound muscle action potential (CMAP) amplitudes. In the upper limbs, median motor nerve response from APB was abnormal in 71% of nerves tested (74% of patients), Ulnar motor response to abductor digiti mini was abnormal in 38% and to first dorsal interosseous in 71% of nerves. Terminal motor latency was significantly reduced in nerves with reduced CMAP amplitude compared with those of normal amplitude in each of the motor nerves. Tibial, median and ulnar nerves with reduced CMAP amplitude demonstrated significantly reduced persistence of F-waves when compared with normal nerves, while minimum F-wave latencies were significantly prolonged in tibial and median nerves with reduced CMAP amplitudes. Sensory nerve conduction values were abnormal in 11% of patients, most commonly with findings consistent with mild axonal sensory neuropathy. A total of 672 muscles were sampled using standard electromyographic techniques. Combined active and chronic denervation were detected in 10.6-65.4% of sampled muscles, with highest rates in APB, first dorsal interosseous and tibialis anterior muscles. Rates of detection of isolated chronic denervation ranged from 62.7-80.4%. The patterns of nerve conduction and electromyographic abnormalities were analysed in relation to the region of onset, and variation in the pattern of abnormalities was seen.

Discussion: The patterns of electromyographic abnormality described in this study may assist in planning an electrodiagnostic study based on the region of clinical onset in order to maximise the rate of detection of neurogenic change. Additionally, selective large motor neuron loss was suggested by prolongation of distal motor and minimum F-wave latencies in nerves with neurophysiological evidence of axonal loss.

P65 PROGRESSIVE AXONAL DYSFUNCTION UNDERLIES CLINICAL IMPAIRMENT IN ALS

CHEAH B^1,3, LIN C^1,3, PARK S^1,3, VUCIC S^1,2, KRISHNAN A¹, KIERNAN M^1,3

¹Neuroscience Research Australia, Sydney, New South Wales, Australia, ²Sydney Medical School, Westmead, New South Wales, Australia, ³Prince of Wales Clinical School, Sydney, New South Wales, Australia

Email address for correspondence: [email protected]

Keywords: clinical trials, axonal excitability, biomarker

Axonal excitability studies undertaken in amyotrophic lateral sclerosis (ALS) patients have implicated Na + and K + channel dysfunction. It remains unknown, however, as to how these changes evolve longitudinally, and whether these changes are related to pathogenesis, neurodegeneration and clinical features. The objectives of the present study were to investigate longitudinal changes in axonal excitability and function in ALS patients and their relationship with axonal loss and clinical impairment. Axonal excitability studies were undertaken in 37 ALS patients (22 males; mean age, 53.5 years; SD, 10.2 years; median disease duration, 12.6 months; inter-quartile range, 9.5 - 26.4 months) at baseline and at 12 weeks follow-up. Measurements at baseline were compared to 39 healthy control subjects (mean age, 52.0 years; SD, 12.55 years). Functional impairment was evaluated using the ALS Functional Rating Scale-revised (ALS FRS-r). Patients were subdivided according to two mutually exclusive classification systems: 1) 10% reduction in peak compound muscle action potential amplitude; and 2) decline in fine motor subscore of ALS FRS-r. Baseline strength-duration time constant (a measure of nodal persistent Na + conductance) was significantly prolonged in ALS patients (mean, 0.48ms; SD, 0.11ms) compared to healthy controls (mean, 0.44ms; SD, 0.11ms; P = 0.02), indicating increased entry of Na + into the axons of ALS patients. By contrast, multiple measures of internodal K + channel function were signicantly reduced at baseline. Longitudinal changes in all excitability variables implicated membrane hyperpolarisation as the primary phenomenon evolving in ALS patients. Furthermore, patients with preserved peak compound motor action potential amplitude demonstrated more severe changes in axonal excitability than those in whom compound potentials declined. Patients who reported progression in fine motor functional impairment generally recorded more severe changes in axonal excitability than those who did not experience decrement in fine motor function. Abnormalities in axonal function at baseline were consistent with increased nodal persistent Na + conductances in ALS patients. Longitudinal studies suggested that peripheral nerves of ALS patients were becoming progressively hyperpolarized. Axonal hyperpolarization may be a manifestation of increased firing of surviving motor units, compensating for neurogenic weakness.

P66 NEUROPHYSIOLOGICAL INDEX: EXPERIENCE WITH 142 ALS PATIENTS

DE CARVALHO M^1,2, PINTO S¹, SWASH M^1,3

¹Neuromuscular Unit, Instituto de Medicina Molecular; Institute of Physiology, Faculty of Medicine, Lisbon, Portugal, ²Department of Neurosciences, Hospital de Santa Maria, Lisbon, Portugal, ³Departments of Neurology and Neuroscience, Royal London Hospital, Queen Mary University of London, London, UK

Email address for correspondence: [email protected]

Keywords: neurophysiological index, M-wave amplitude, motor unit number estimation

Introduction: The neurophysiological index (NI) was described in 2000 by de Carvalho and Swash, as a simple and sensitive marker to measure lower motor (LMN) loss over ALS progression. Since then, other groups have applied this tool, in particular in clinical trials. We aim to review our experience on applying this marker in 142 patients with ALS from 2003 to 2010.

Methods: We analysed the mean percentage of change per months (%/m) for ALS-FRS, M-wave amplitude (ampl), MUNE and NI. MUNE was tested in 142 patients, 100 of whom with incremental technique (MUNE incr) and the last 24 with multipoint stimulation. This summarizes results published in 2003, 2005, 2006 and 2010, as well as new data (26 patients). The global results were adjusted for the number of patients included in each study.

Results: The mean age and gender distribution were not significantly different among studies, but disease duration showed large variation (mean of 22, 18, 16.4, 13 and 12 months, for studies 1 to 5, respectively). From study 1 to 5 (18, 39, 33, 28 and 24 patients, respectively), the mean change (%/m) was the following for ALS-FRS (3.57; 2.60; 2.02; 3.07; 2.90); ampl (3.42; 2.78; 5.04; 4.27; 3.22); NI (6.08; 4.48; 7.41; 6.99; 5.79); MUNE (not done; 4.70; 7.58; 6.74; 4.47). The variability (maximal - minimum value/mean) among results from different populations of ALS patients were 0.54, 0.60, 0.48, 0.53 (0.45 for MUNE incr) for ALS-FRS, ampl, NI and MUNE, respectively. We did not find any significant correlation between disease duration and rate of decline for the different measurements (Spearman's rho), or between ALS-FRS and the neurophysiological changes. The only significant correlation observed was between ampl and NI (p < 0.001).

Discussion: There is a small variation between among different groups of ALS patients regarding the rate of change of ALS-FRS, ampl, NI and MUNE, but NI and MUNE showed the smaller inter-group variation, indicating a good consistency of these markers. There was no correlation between disease duration and rate of change, suggesting that disease progression is not only determined by this factor. The strong correlation between ampl and NI represents that NI is derived from ampl. Our results support NI as a simple, sensitive and consistent marker of disease duration.

P67 COMPARISON OF MOTOR UNIT NUMBER ESTIMATION BY BAYESIAN STATISTICAL MUNE WITH HISTOLOGICAL COUNTING OF MOTOR UNIT NUMBERS: VALIDATION OF THE METHOD

NGO S¹, BELLINGHAM M², PETTITT A³, RIDALL G³, HENDERSON R⁴, MCCOMBE P^1,4

¹The University of Queensland Centre for Clinical Research, Brisbane, Queensland, Australia, ²The University of Queensland, Brisbane, Queensland, Australia, ³Queensland University of Technology, Brisbane, Queensland, Australia, ⁴Royal Brisbane and Women's Hospital, Brisbane, Queensland, Australia

Email address for correspondence: [email protected]

Keywords: MUNE, neurophysiology, biomarkers

Background: The number of motor units innervating a single muscle or a group of muscles determines muscle function. Motor unit number estimation techniques are commonly used to determine motor unit loss in ALS. We developed a method of motor unit number estimation, using neurophysiological data collected as the stimulus-response curve and Bayesian statistical analysis. Because this technique cannot be validated in humans, we tested this technique in a mouse model of ALS to compare the MUNE results with histological counts.

Objectives: Determine whether the number of motor units determined by Bayesian statistical MUNE is an accurate representation of the number motor units in a muscle, by comparing MUNE results with histological counts of motor neurons.

Methods: Male and female wild-type and SOD1G93A transgenic mice were studied at three stages of disease. Stimulus-response curves were obtained by application of a graded stimulus to the right sciatic nerve and collection of the compound muscle action potential from the right gastrocnemius muscle. Spinal cords were collected and motor neurones in the L4-5 spinal cord were counted by a physical fractionator method. Motor unit number was also calculated by our Bayesian MUNE method.

Results: In 10 presymptomatic wild-type animals, MUNE ranged from 40-54 and the number of motor units with histology ranged from 61-94. In 10 presymptomatic SOD1G93A mice, MUNE ranged from 27-53 and the number of units with histology ranged form 44-81. In 10 wild-type animals at disease onset age, MUNE ranged from 23-50 and histological counts ranged between 47-74. Onset SOD1G93A mice had a MUNE of 11-35 with histological counts ranging from 31-47. Ten wild-type animals at the end-stage age had MUNE ranging from 29-48 and histological counts from 53-97. By the end-stage of disease, SOD1G93A animals had much fewer motor units. MUNE ranged from 4-15 and histological counts ranged between 18-30 on 10 SOD1G93A end-stage animals. In all animals, the histology to MUNE ratio was approximately 2:1. Histological counts were highly correlated to MUNE (p = 0.0168 for wild-type and p < 0.0001 for SOD1). The progressive loss of motor units in SOD1G93A mice during disease progression occurred concurrently with denervation at the neuromuscular junction (p < 0.001 for all groups, n = 5 per group, t-test). There were no significant differences noted between male and female SOD1G93A mice.

Discussion and conclusions: Our MUNE method gives results that are approximately half of the number of anterior horn cells counted with histology. As our physiological method only measures functioning motor units, it is expected to be less than the anatomical count, which includes both functional and non-functional anterior horn cells.

P68 PRESYNAPTIC DECLINE OF THE NEUROMUSCULAR TRANSMISSION IN THE SOD1- G93A MOUSE MODEL OF ALS IS PREVENTED BY G-CSF IN GENDER-SPECIFIC MANNER

POLLARI E, NAUMENKO N, KURRONEN A, GINIATULLINA R, SHAKIRZYANOVA A, MAGGA J, GINIATULLIN R, KOISTINAHO J

Department of Neurobiology, A. I. Virtanen Institute for Molecular Medicine, University of Eastern Finland, Kuopio, Finland

Email address for correspondence: [email protected]

Keywords: G-CSF, electrophysiology, neuromuscular junction

Background: Epidemiological studies show that both the incidence and prevalence of ALS are greater in men than in women. Sex differences have also been reported in mice overexpressing mutant SOD1, including the delayed onset in female G93A-SOD1 mice. However, the mechanisms underlying the role of gender differences in ALS are unknown. Because oxidative stress may be involved in motoneuron death in ALS and synapses are thought to be sensitive to oxidative stress, we characterized electrophysiologically the synaptic function in male and female G93A-SOD1 mice at early symptomatic stage of the disease.

Objectives: Our aim was to measure the synaptic properties of neuromuscular junction in diaphragm of an ALS mouse model and determine whether sex or anti-inflammatory treatment with granulocyte colony stimulating factor (G-CSF) affects the properties of synaptic transmission or level of reactive oxygen species (ROS) in spinal cord.

Methods: Transgenic mice overexpressing mutant Cu, Zn superoxide dismutase, G93A-SOD1, received weekly s.c. injections of G-CSF with sustained action, starting at pre-symptomatic stage. Properties of synaptic transmission and amount of ROS were measured from ALS mice at early symptomatic stage. Using intracellular microelectrode technique miniature and evoked endplate potentials (MEPPs and EPPs) were recorded in the diaphragm muscle and the amount of ROS was measured from the lumbar spinal cord.

Results: Electrophysiological testing revealed that the postsynaptic function was mainly preserved in G93A-SOD1 mice whereas the presynaptic properties were greatly affected by the disease. At postsynaptic site the membrane potential of muscle fibres was unaffected and postsynaptic sensitivity, measured as amplitude of MEPPs, was only slightly reduced in G93A-SOD1 mice. At presynaptic site, MEPPs were present in neuromuscular synapses of both sexes in wild type and transgenic mice. However, in transgenic mice the probability of spontaneous release was lowered and readily releasable transmitter pool was reduced when compared to those of wild type mice. In male G93A-SOD1 mice ROS reduced spontaneous quantal release, while female mice were resistant to this inhibitory action of ROS. The disease inflicted synaptic dysfunctions were improved by G-CSF treatment in male mice whereas the effect of the treatment was minor in females. Similarly, in the spinal cords of the G93A-SOD1 mice the level of ROS was increased in males but not in females and was reduced by G-CSF treatment.

Conclusions and discussion: This is the first detailed electrophysiological analysis of impaired synaptic function in a mouse model of ALS. In addition to the characterization of neuromuscular transmission mechanisms, we observed gender-specific prevention of synaptic impairment with anti-inflammatory G-CSF treatment.

P69 COL19A1 AS A POTENTIAL PROGNOSTIC BIOMARKER IN ALS

CALVO A¹, ATENCIA G², JUÁREZ-RUFIÁN A², CORDERO VÁZQUEZ P², TORRE MERINO P², MARTÍN MA³, MANZANO R¹, GASCO S¹, OLIVÁN S¹, MUÑOZ BLANCO JL⁴, ESTEBAN PÉREZ J², MUÑOZ MJ¹, ZARAGOZA P¹, GARCÍA REDONDO A², OSTA R¹

¹LAGENBIO-I3A, Aragonese Institute of Health Sciences (IACS), Faculty of Veterinary, University of Zaragoza, Zaragoza, Spain, ²Neurology Department – ALS Unit, Ciberer U-723, Health Research Institute, October 12th University Hospital, Madrid, Spain, ³Biochemistry Department, Ciberer U-723. Health Research Institute, October 12th University Hospital, Madrid, Spain, ⁴Neurology Department – ALS Unit, Gregorio Marañón Hospital, Madrid, Spain

Email address for correspondence: [email protected]

Keywords: biomarker, Col19a1, muscle biopsy

Background: How skeletal muscle degenerates in ALS is currently unknown. However this tissue is one of the most promising targets for the search of potential biomarkers of ALS (1,2). In particular, one of the genes that is involved in regenerative pathways, to maintain the integrity of the tissue, is collagen type XIX, alpha 1 (Col19a1) (3). Due to the fact that regenerative processes tend to compensate for the induced unbalance under ALS degeneration, the possible role of Col19a1 as a prognostic biomarker of ALS is under consideration.

Objectives: Our main goal is to study the transcriptional expression levels of Col19a1 in skeletal muscle of transgenic SOD1^G93A mice and in biopsy samples from ALS patients in order to analyze its role as a potential prognostic biomarker of longevity.

Methods: Muscle biopsies were carried out in forty eight SOD1^G93A mice at each disease stage. Three biopsy samples were obtained per animal. Muscle biopsies from 8 ALS patients and 4 controls were obtained with prior informed consent. Gene expression variations in all samples were assayed by real-time PCR. Pearson correlation coefficient and Wilcoxon test were used in the statistical analysis.

Results: The longevity range in transgenic SOD1^G93A varied from 105 to 160 days. The transcriptional levels of Col19a1 varied significantly along disease progression and also correlated positively with longevity in SOD1^G93A mice. The mean age in ALS patients (7 men and a woman) was 58.26 ± 5.59 years and 77 ± 6.4 years in controls (4 women). The transcriptional expression of Col19a1 in ALS patients was significantly higher than in controls.

Discussion and conclusions: The significantly different gene expression profile and correlations of Col19a1 found in transgenic SOD1^G93A and in muscle biopsies of ALS patients and controls suggested that Col19a1 could be considered as a potential prognostic biomarker of longevity in ALS. In neurodegenerative conditions such as ALS, Col19a1 could compensate for the activation of degeneration signals in skeletal muscle, prompting muscle differentiation and regeneration. These results could shed light to find new biomarkers of ALS and finally a more accurate knowledge of the disease could be achieved.

Acknowledgements: This work was supported by grants PI071133 and PI060201 from the Fondo de Investigación Sanitaria of Spain, grant SAF2009-12495 from MICIN Spain, grants PIPAMER 08/08 and 09/09 from the Aragon Institute of Health Sciences, the Project “Tú eliges: tú decides” of Caja de Ahorros de Navarra in Spain and by ISCIII (PI071283 and EC08/00049).

Reference

• Dobrowolny G, Aucello M, Rizzuto E, et al. Cell Metabolism 2008;8:425–36.
   PubMed Web of Science ®Google Scholar
• Turner MR, Kiernan MC, Nigel P, et al. The Lancet 2009;8:94–109.
   Google Scholar
• Sumiyoshi H, Mor N, Lee SY, et al. J. Cell. Biol. 2004;166:591–600.
   PubMed Web of Science ®Google Scholar

P70 NEUROTROPHIN RECEPTOR P75 AS A BIOMARKER FOR MOTOR NEURON DISEASE

SHEPHEARD S, CHATAWAY T, RUSH R, ROGERS M-L

Flinders University, Flinders Medical Science & Technology, Human Physiology and Centre for Neuroscience, Adelaide, South Australia, Australia

Email address for correspondence: [email protected]

Keywords: biomarker, neurotrophin receptor p75 (p75NTR)

Background: There are no biochemical biomarkers for Motor Neuron Disease (MND), thus, finding a biological measure of the extent of this disease is critical. The American National Institute of Health defines a biomarker as ‘a characteristic that can be measured and evaluated as an indicator of normal biological processes, pathological processes or pharmalogical responses to therapeutic intervention’ (1). An important step in finding effective treatments is to identify biomarkers that could aid in the detection and progression of disease, and more rapid translation of potential therapeutics from models to clinical trials. The neurotrophin receptor p75 (p75NTR) is highly expressed during development and is greatly reduced in different types of cells in adulthood (2). However, multiple studies have found that p75NTR is robustly induced by injury (3). Of particular interest, is that p75NTR has been found upregulated in the spinal cord of: persons with MND post-mortem (4) and the SOD1G93A mouse model of MND (5), and also in urine following sciatic nerve injury in rats (2).

Objectives: To determine if p75NTR is detectable in the urine of the MND mouse model SOD1G93A, and whether it could be used as a biomarker in the mouse model, allowing for more rapid translation of possible therapeutics to clinical trials.

Methods: SOD1G93A behavioural analysis was performed using neurological assessment, weight and hanging wire tests. The presence of p75NTR in SOD1G93A mice urine was detected with Western blot and confirmed by Immunoprecipitation.

Results: Behavioural testing showed that SOD1G93A mice (n = 10) developed motor symptoms at 120 days of age reaching end-stage by 150 days. p75NTR was detectable in SOD1G93A mice urine as early as 60 days of age, increasing with age, but was not detectable in age-matched controls (n = 10) until after 120 days.

Discussion: p75NTR was detectable pre-symptomatically in SOD1G93A mouse urine and may be useful as a biomarker of disease in the SOD1G93A model to allow for more rapid translation of potential therapeutics from the model to human clinical trials.

Conclusion: p75NTR has been identified as a possible biomarker for MND in the SOD1G93A mouse and on-going work is being undertaken to determine if it could be of use as a biomarker in human disease. Blood and urine samples from SOD1G93A mice and persons with MND will be tested for p75NTR by developing a more sensitive, quantitative ELISA assay.

Reference

• Atkinson A, Colbourn W, Degruttola V, et al. Clinical Pharmacology & Therapeutics 2001;69:89–95.
   PubMed Web of Science ®Google Scholar
• DiStefano P, Johnson E. Proceedings of the National Academy of Science 1988;85:25–40.
   Google Scholar
• Hennigan A, O’Callaghan R, Kelly A. Biochemical Society Transactions 2007;35:424–27.
   PubMed Web of Science ®Google Scholar
• Seeburger J, Tarras S, Natter H, et al. Brain Research 1993;621:111–5.
   PubMed Web of Science ®Google Scholar
• Copray J, Jaarsma D, Ku B, et al. Neuroscience 2003; 116:685–94.
   PubMed Web of Science ®Google Scholar

P71 DECREASED URINARY CONCENTRATIONS OF TYPE IV COLLAGEN IN AMYOTROPHIC LATERAL SCLEROSIS

MIKAMI H, ONO S

Teikyo University Chiba Medical Center, Ichihara, Chiba, Japan

Email address for correspondence: [email protected]

Keywords: urine, serum, type IV collagen

Background: Recently, we have demonstrated that the basement membrane of skin in ALS patients was weakly positive for type IV collagen (IV-C) as compared with that of controls and that serum IV-C levels in patients with amyotrophic lateral sclerosis (ALS) were lower than in controls, suggesting that a metabolic alteration of IV-C may take place in the skin of ALS and that the decreased level of serum IV-C may reflect the decreased IV-C immunoreactivity of skin in ALS. We have postulated that measurement of urinary IV-C levels could be useful in assessing the alterations in basement membrane IV-C metabolism in the skin in ALS and for monitoring the progression of ALS.

Objectives: The aim of the present study was to measure the urinary concentrations of IV-C.

Methods: Our subjects were 20 patients with ALS(mean age ± SD, 57.6 ± 7.2 years; 14 men and 6 women), 20 diseased control subjects with other neurologic and muscular diseases (58.2 ± 9.1 years; 13 men and 7 women), and 20 age- and sex- mached healthy adults (51.4 ± 9.1 years, 12 men and 8 women). A polyethylene glycol (PEG)-based concentration method was used for concentrating urinary IV-C. Urinary IV-C concentrations were divided by urinary concentrations of creatinine to exclude the influence of various concentrations in urine. Urinary IV-C concentrations were measured by a one-step sandwich enzyme immunoassay (EIA).

Results: Urinary IV-C levels were 3.0 ± 1.5 ng/ml, 5.1 ± 2.0 ng/ml, and 5.7 ± 2.0 ng/ml in ALS patients, diseased control subjects, and healthy control subjects, respectively. The urinary level of IV-C was significantly lower in ALS patients than in diseased control subjects (p < 0.001) and healthy controls (p < 0.001). There was no appreciable difference in the urinary IV-C level between diseased control subjects and healthy controls. The urinary IV-C level was significantly and negatively correlated with duration of symptoms in patients with ALS (r = -0.85, p < 0.001), but there was no such correlation in diseased control subjects. No correlation was found between urinary concentrations of IV-C and dysphagia, muscle power rating (r = -0.28, 0.1 < p < 0.5), severity of disability (r = 0.30, 0.1 < p < 0.5) in ALS patients or diseased control subjects.

Discussion and conclusions: We used a 1-step sandwich EIA to measure urinary IV-C. Because this assay uses antibodies against both the 7S portion and the central triple helical portion, urinary IV-C measured by a sandwich EIA is a whole molecule and not a degradation product of preexisting IV-C. Measurement of decreased levels of urinary IV-C by this method probably reflects its decreased synthesis or increased degradation. Urinary concentrations of IV-C may be a sensitive indicator of altered IV-C metabolism and could also reflect basement membrane metabolism. These data suggest that a metabolic alteration of IV-C may occur in ALS patients.

P72 ELEVATED SERUM LEVELS OF AUTOANTIBODIES AGAINST HEAT SHOCK PROTEINS IN PATIENTS WITH AMYOTROPHIC LATERAL SCLEROSIS

HWANG CS^1,2, TSAI PW¹, CHANG DT¹, CHANG HT³

¹Institute of Molecular and Cellular Biology & Department of Medical Science, National Tsing Hua University, Hsinchu, Taiwan, ²Taipei City Hospital - Zhong Xiao Branch, Taipei, Taiwan, ³Graduate Institute of Molecular Systems Biomedicine, China Medical University, Taichung, Taiwan

Email address for correspondence: [email protected]

Keywords: autoantibody, heat shock proteins, biomarker

Background: Nowadays, the diagnosis of amyotrophic lateral sclerosis (ALS) is still mainly based on clinical symptoms and electromyographic findings. Hence a high rate of misdiagnosis and delays in proper diagnosis are almost unavoidable. Recent studies using proteomic approaches have identified a variety of potential serum biomarkers for the diagnosis of ALS. However, none of these has been proven to be effective for clinical use and further research is necessary.

Objectives: The expressions of heat shock proteins (HSPs) have been reported to be decreased before symptom onset and increased after symptom onset in brain tissue of G93A mice. These phenomena could be applied to develop potential biomarkers for ALS. HSPs belong to the damage associated molecular patterns (DAMPs), which are released by damaged cells, and could bind Toll-like receptors to activate inflammatory responses. To reduce inflammation, the immune system will produce autoantibodies to neutralize the inflammatory ligands, including HSPs. The aim of this study was to determine whether serum autoantibodies against HSPs could serve as surrogate biomarkers for the clinical diagnosis of ALS.

Methods: Forty ALS patients and 40 age-matched healthy adults were recruited. The ages of the patients were 32 to 86 y/o (mean = 61, SEM = 13), and that of the controls were 30 to 87 y/o (mean = 60, SEM = 17). The clinical severity of ALS was evaluated using the revised ALS functional rating scales (ALSFRS-R). The ALSFRS-R scores of our patients ranged from 0 to 33 (mean = 9.18, SEM = 10.22), and the disease duration ranged from 11.5 to 142.4 months (mean = 60.7, SEM = 38.5). The sera of patients and controlled subjects were collected. The serum autoantibodies against HSP60 and HSP70 were measured using enzyme linked-absorbent assay (ELISA). Student t-test was employed for statistics.

Results: Our results showed that the levels of serum autoantibodies against HSP60 and HSP70 were significantly higher (1.64- and 1.37-fold higher) in ALS patients as compared with those of age-matched controls. For the diagnosis of ALS, the performance of HSP60 autoantibodies showed 73% sensitivity, 74% specificity and 72% accuracy, and that of HSP70 autoantibodies showed 60% sensitivity, 60% specificity and 60% accuracy. The AUC values were 0.7651 and 0.6179 for autoantibodies against HSP60 and HSP70, respectively. Meanwhile, there is a tendency of increasing serum HSP60 and HSP70 autoantibody levels as disease severity increasing.

Conclusions: The elevated levels of HSPs have been reported in various neuronal injuries and neurodegenerative diseases. Our study provides a new idea that serum autoantibodies against HSP60 and HSP70 may be novel surrogate biomarkers for the diagnosis of ALS. This discovery also provided a clue to study the roles of serum autoantibodies in ALS.

P73 SERUM CREATININE (CR) TO CYSTATIN C (CYSC) RATIO (CR/CYSC) IS A GOOD SURROGATE MARKER OF RESIDUAL MUSCLE VOLUME IN AMYOTROPHIC LATERAL SCLEROSIS

TETSUKA S-U, AKIMOTO C, MORITA M, KAWAKAMI T, IKEGUCHI K, NAKANO I

Jichi Medical University, Simotsuke City, Tochigi Prefecture, Japan

Email address for correspondence: [email protected]

Keywords: serum creatinine, serum cystatin C, surrogate marker

Background: The severity grade of ALS is approximately associated with the residual volume of skeletal muscles, and a parameter that reflects the muscle volume would enable us to estimate the severity. Since serum creatinine (Cr), though almost exclusively originating from skeletal muscle, is dependent on renal function, it cannot be such a marker in ALS patients. Cystatin (CysC), which is derived from all nucleated cells and independent of the body muscle volume, is excreted from the kidneys in the same way as Cr. Thus, the Cr/CysC ratio, which remains almost constant irrespective of the renal function in individuals without neuromuscular diseases, is supposed to be a good biomarker of the muscle volume, inverse of ALS severity.

Objective: To determine whether or not serum Cr/CysC may serve as a good surrogate marker of residual muscle volume, i.e., the inverse of the severity of ALS.

Methods: The serum levels of Cr and CysC were measured in 62 ALS patients admitted to our faculty during 2008-2010 (39 men, 23 women; mean age ± SD (years) = 62.9 ± 9.6) and 50 control subjects (mean age ± SD (years) = 62.8 ± 12.8). The correlation between Cr/CysC and severity grade of ALS (grades 1, 2, 3, 4 and 5 in ascending order of severity; the Research Committee of CNS Degenerative Diseases, the Ministry of Health, Labor and Welfare of Japan) was assessed.

Results: The Cr/CysC ratio was 7.90 ± 2.22 in the controls and 5.47 ± 3.31 in the ALS patients (p < 0.05). In ALS patients, the means of serum Cr were 10.13, 7.82, 6.47, 6.30, and 2.37 (F (4,56) = 18.19, p < 0.001) for ALS severity grades 1, 2, 3, 4 and 5, Japanese people respectively. Meanwhile, the serum Cr/CysC are 1.01, 0.78, 0.64, 0.63, and 0.23 for the each severity grade respectively (F(4,57) = 19.99, p < 0.001). Post-hoc test with Tukey's HSD, we revealed significant differences in Cr/CysC between severity grades 1-4, 1-5, 2-5, 3-5, and 4-5 (p < 0.05).

Discussion and conclusions: In our study, Cr/CysC in ALS showed a clear inverse correlation with the disease severity, giving rise the possibility that it could be a renofunction-independent surrogate marker for assessment of the disease severity and progression. The ratio can be applied to various clinical trials in this condition.

P74 IDENTIFICATION OF A POTENTIAL SPORADIC ALS BIOMARKER IN CEREBROSPINAL FLUID

MACKLE T^1,3, PELECH S^1,2, SHAW CA^1,3

¹University of British Columbia, Vancouver, BC, Canada, ²Kinexus Bioinformatics Corporation, Vancouver, BC, Canada, ³Neural Dynamics Research Group, Vancouver, BC, Canada

Email address for correspondence: [email protected]

Keywords: biomarkers, cerebrospinal fluid, Eph receptors

Background: The current lack of disease markers for sporadic ALS (sALS) has negative consequences for patients when it comes to clinical diagnosis, early medical intervention and participation in therapeutic trials. Valid biomarkers can be useful for diagnostic and prognostic indications as well as providing insight into disease pathogenesis and identifying targets for therapeutic interventions. Cerebrospinal fluid (CSF) may be a particularly valuable source of biomarkers because it is in close anatomical proximity to the brain and spinal cord, thus making it a better reflection of biochemical alterations resulting from neurodegenerative disease.

Objectives: We examined the proteome profile of CSF from sALS patients and controls to identify candidate biomarkers that could aid in the diagnosis of ALS and possibly provide insight into disease pathogenesis.

Methods: An antibody microarray technique was used to measure the expression levels of various cell signalling proteins in pooled CSF samples from sALS patients and healthy controls. Follow-up analysis by Western blotting with CSF from a separate cohort of sALS patients (n = 19), healthy controls (n = 21) and neurological disease controls (n = 10) were performed in order to validate protein changes observed in the microarray.

Results: Initial proteomic discovery studies revealed a decrease in the level of ephrin type-A receptor 1 tyrosine kinase (EphA1) in the pooled CSF of sALS patients relative to healthy controls. Independent validation studies with a separate cohort of sALS, healthy control and neurological disease control subjects confirmed that EphA1 expression in sALS CSF was significantly lower than both control groups. As a diagnostic test, EphA1 levels in CSF had a statistically significant ability to discriminate between sALS patients and controls. EphA1 levels also had a positive correlation with age at onset of sALS symptoms, indicating that this potential biomarker may be capable of identifying people who are prone to an earlier onset of disease.

Discussion and conclusions: The Eph family of tyrosine kinase receptors engage in complex bidirectional signalling that functions as a major form of contact-dependent communication between cells, and mediates a variety of cellular responses. As research begins to further elucidate these multifaceted signalling mechanisms, imbalances in Eph receptor functioning are increasingly implicated in a variety of pathologies in the central nervous system that are related to some of the major features of ALS. Larger scale studies are required to verify and further validate the candidate EphA1 biomarker, as well as define the functional implications of its down-regulation in sALS CSF. EphA1 should also be tested in conjunction with other potential CSF biomarkers to see if a more robust diagnostic test for ALS can be achieved. Finally, the levels of other receptors in the Eph family should be investigated and possible correlations with clinical parameters explored.

P75 BIOCHEMICAL ALTERATIONS ASSOCIATED WITH ALS

BERRY JD¹, CUDKOWICZ ME¹, BROWN MV³, ALEXANDER D³, CAFFREY R³, WULFF JE³, BOWSER RT², LAWSON RJ¹, JAFFA M¹, MILBURN MV³, RYALS JA³, LAWTON KA³

¹Massachusetts General Hospital, Boston, MA, USA, ²University of Pittsburgh, Pittsburgh, MA, USA, ³Metabolon, Inc., Durham, NC, USA

Email address for correspondence: [email protected]

Keywords: biomarker, metabolomics, pathophysiology

Background: There is no single diagnostic test for amyotrophic lateral sclerosis (ALS); imaging and laboratory tests are used to rule out diseases mimicking ALS. Electromyography can support an ALS diagnosis but is not specific. Average duration between symptom onset and diagnosis is 1 year. Metabolomics, the study of small molecules in biological systems is a potentially powerful technique to evaluate global biochemical alterations in a variety of diseases. Metabolomics could provide an understanding of the biochemical basis of ALS and identify biomarkers for diagnosis and disease progression. This could facilitate early diagnosis and trial entry, provide surrogate outcome measures, and decrease trial duration and size.

Objective: The plasma of ALS patients and healthy volunteers was analyzed for changes in the metabolic pathways in ALS patients manifesting as a uniquely altered metabolic signature.

Methods: Two cross-sectional biofluid collection studies were conducted. Plasma was collected from 62 (Study 1) and 99 (Study 2) ALS patients meeting El Escorial criteria for possible, probable, or definite ALS; 69 (Study 1) and 48 (Study 2) samples were collected from healthy volunteers. Unbiased metabolomic analysis was carried out on three independent instrument platforms: one gas chromatography/mass spectrometry (GC/MS) and two previously described ultrahigh performance liquid chromatography/tandem mass spectrometry (UHLC/MS/MS2) platforms optimized for either basic species or acidic species (1,2). Welch's two-sample t-test was employed to identify biochemicals with altered levels in ALS patients relative to healthy volunteer groups. For this hypothesis-generating analysis, biochemicals where p < 0.05 and q < 0.2 were considered statistically significant.

Results: A total of 335 biochemicals were identified in the plasma samples of the two studies. Of these metabolites, 240 (72%) were measured in both studies. In Study 1, ninety-two out of 282 (33%) metabolites were significantly altered in ALS patients, and in Study 2, fifty-five out of 293 (19%) were significantly altered. Twenty-three metabolites were significantly increased or decreased in both studies.

Discussion and conclusions: Using metabolomics, specific changes in the metabolic profile can be identified in patients with ALS relative to healthy volunteers. These changes point to potential disease mechanisms and potential therapeutic targets. The data presented here may provide insight into the pathophysiology of ALS and suggest promising areas for future studies. Metabolites involved in proposed ALS disease mechanisms were identified in the present analysis, including hypermetabolism, mitochondrial dysfunction, oxidative damage, hepatic dysfunction and neuronal breakdown.

Reference

• Evans AM, DeHaven CD, Barrett T, et al. Annal Chem 2009;81:6656–67.
   PubMed Web of Science ®Google Scholar
• Ohta T, Masutomi N, Tsutsui N, et al. Toxicol Pathol 2009;37:521–35.
   PubMed Web of Science ®Google Scholar

P76 BIOMARKER PROFILE DIFFERS BETWEEN CLINICALLY DEFINITE PRIMARY LATERAL SCLEROSIS (PLS) AND AMYOTROPHIC LATERAL SCLEROSIS (ALS)

MITSUMOTO H¹, SANTELLA R M¹, CREMERS S¹, PASMANTIER M¹, ARMSTRONG N¹, SHOESMITH C², FLOETER MK³, BAROHN R⁴, SINGER M⁵, WOLFE G⁵, CAMPANELLA C¹, GINSBERG HN¹, FACTOR-LITVAK P¹, STUDY GROUP ALS/PLS COSMOS¹

¹Columbia University, New York, NY, USA, ²University of Western Ontario, London, Ontario, Canada, ³National Institutes of Health, Bethesda, MD, USA, ⁴University of Kansas, Kansas City, KS, USA, ⁵University of Texas Southwestern, Dallas, TX, USA

Email address for correspondence: [email protected]

Keywords: PLS, biomarkers, oxidative stress

Background: PLS is clinically characterized by pure upper motor neuron dysfunction (PUMND) from onset throughout the disease course. Such criteria require that a diagnosis of PLS should not be made until at least 36 to 48 months after symptom onset. While long-term survival in PLS is common, it causes marked, progressive motor disability. To date, it is unknown if PLS is an independent disease or part of the MND spectrum. Recent neuroimaging and neurophysiological findings show unique features of PLS, but whether there are distinctive biochemical biomarker differences remains unknown.

Objectives: To determine if there are differences in oxidative stress (OS) biomarkers and lipid profiles in patients with PLS compared to those with ALS.

Methods: Overnight fasting urine and plasma specimens were collected from patients with clinically definite PLS (disease duration of at least 5 years but < 15 years and normal EMG done within 1 year of enrollment) in our on-going Spastic Paraplegia Foundation Multicenter PLS Study. Age- (+ 5 years) and sex-matched biosamples from patients with ALS (based on the El Escorial criteria and less than 18 months after symptom onset) were selected from the ALS Cohort Study of Multicenter Oxidative Stress (COSMOS) patient biobank. Lipid profiles were analyzed by standard lipid analyses (n = 13 both in ALS and PLS) and urinary OS markers were measured by immunoassay (n = 15 both in ALS and PLS).

Results: Total cholesterol and triglyceride levels were slightly higher in PLS (190 ± 18 and 129 ± 21 mg/dl, respectively) than ALS (151 ± 14 and 109 ± 16 mg/dl) but not statistically significant (p = 0.1 and 0.4, respectively). LDL was significantly higher in PLS (124 ± 14 mg/dl) than ALS (88 ± 10 mg/dl; p = 0.05). The ratio of LDL/HDL was higher in PLS (3.2 + 1.4) than ALS (2.2 + 0.9; p = 0.06). An OS biomarker, urinary isoprostanes (lipid oxidation products), was not different, but levels of urinary 8-oxodeoxyguanosine (oxidized DNA base) were significantly elevated in ALS (28 + 11 nmol/mmol) compared to PLS (21 + 9 nmol/mmol; p = 0.06).

Discussion and conclusions: Our study suggests some biochemical differences between PLS and ALS despite the small sample size. An increased ratio of LDL/HDL has been associated with longer survival or with less respiratory distress in patients with ALS. Our findings are consistent with the less respiratory distress and longer survival of PLS patients. Further, the findings suggest lower levels of oxidative stress among PLS cases. This is the first demonstration of biochemical differences between PLS and ALS. Further studies are clearly needed to elucidate the progression of biomarkers in the natural history of PLS. (The study is funded by R01-ES016348, MDA, and Spastic Paraplegia Foundation).

P77 TOWARDS DEVELOPING A LONGITUDINAL WHOLE-BRAIN PROTON-MR METABOLITE IMAGING DATABASE FOR EVALUATING TEMPORAL CHANGES OF METABOLITES IN ALS

SHARMA KR, SHERIFF S, SAIGAL G, MAUDSLEY AA, GOVIND V

University of Miami, Miami, USA

Email address for correspondence: [email protected]

Keywords: biomarkers, brain, spectroscopy

Background: Brain metabolite concentrations as measured by proton MRS are known to be altered in several pathologies including ALS. Previous brain-MRS studies in ALS have acquired data from predetermined single-voxel and multi-voxel locations that resulted in assessment of metabolite changes from limited brain regions. The accumulating evidence for extramotor involvement warrants evaluation of metabolite level alterations from the whole brain. In this study, a whole-brain MRSI acquisition and fully-automated data processing and unbiased analyses methods were used to evaluate metabolite changes longitudinally.

Objectives: To develop a multi-time point whole-brain proton-MR metabolite imaging database for evaluating temporal changes in metabolite concentrations and to assess its use for longitudinal studies.

Methods: Seven subjects with definite sporadic-ALS were scanned at 3T, 2-times with intervals of 4-to-10 months for 5 and 19, 29 months for 2 subjects. The whole-brain MRSI were obtained using a volumetric acquisition sequence with TR/TE of 1710/70. The metabolites, N-acetyl aspartate (NAA), total-creatine (Cre), total-choline (Cho) and Cho/NAA were quantified using 43-region and 9-region (8 hemispheric lobes + cerebellum) anatomical atlases. Lobar analysis included calculating values for gray-matter (GM) and white-matter (WM). Comparison of metabolite measures between the 2-times were performed using the 2-sided paired t-test. A p-value of < 0.05 was considered significant.

Results: Using the lobar-atlas, significant metabolite changes were observed in ALS for NAA in the GM of bilateral temporal and left parietal, for Cho in the GM of right temporal, for Cre in the WM of frontal and right temporal lobes. Changes approaching towards significance were observed in left occipital WM for Cre, and left-frontal GM for Cho/NAA. Of the 9 regions, significant changes were observed from 5 regions for at least one of the metabolites and/or tissue-types.

Using the 43-region-atlas, significant metabolite changes were observed in ALS for the following regions: left-occipital, bilateral-parietal, right-paracentral-lobule, bilateral-caudate, bilateral-thalamus and left-temporal for NAA; left-frontal, left-occipital, left-fusiform, right-parietal, right-paracentral lobule, left-caudate, right-pallidum, bilateral-thalamus and bilateral-temporal for Cre; right-postcentral, right-paracentral lobule and right-thalamus for Cho; and right-postcentral for Cho/NAA. The p-values for several other regions showed approach towards significance. In all, 14 of the 43 regions showed significant changes for at least one of the metabolites.

Discussion and conclusions: The observation of significant metabolite changes temporally from several lobar and anatomical substructures of ALS patients indicate that the changes are anatomically widespread. Whole-brain MRSI acquisition method and its unbiased data analyses approach are well-suited for evaluating longitudinal changes in metabolites. Further studies with data from additional number of subjects and time-points are required to assess use of this data for assessing efficacy of drugs.

Funding: NIH grants, R01 NS060874 and R01 EB000822.

P78 WHOLE-BRAIN PROTON MR METABOLITE IMAGING OF ALS AND CORRELATION WITH CLINICAL ASSESSMENTS

GOVIND V, SHARMA KR, SHERIFF S, ARHEART K, SAIGAL G, MAUDSLEY AA

University of Miami, Miami, USA

Email address for correspondence: [email protected]

Keywords: biomarkers, brain, spectroscopy

Background: Brain metabolite changes in ALS were evaluated mostly in the motor cortex (MC) and corticospinal tracts (CSTs) using single-voxel and 2D-MR spectroscopic imaging (MRSI) methods. Such methods are inherently biased as only data from predetermined anatomical regions are acquired. In this study, a whole-brain MRSI acquisition and fully-automated processing and unbiased analyses methods were used to evaluate metabolite changes.

Objectives: To evaluate metabolite alterations in whole-brain MRSI by performing data analyses by tissue-types (white-matter (WM) and gray-matter (GM)), lobes and whole-CSTs, and correlate that with clinical assessments in ALS.

Methods: Thirty-eight subjects with definite sporadic-ALS and 70 age-matched controls were scanned at 3T. Clinical measurements included ALSFRS-R, vital capacity and assessments of upper motor neuron (UMN) function. The whole-brain MRSI were obtained using a volumetric acquisition sequence with TR/TE of 1710/70 ms (1). The metabolites, N-acetyl aspartate (NAA), and Choline (Cho)/NAA were quantified by brain lobes (frontal, temporal, parietal and occipital) and tissue-types and by whole-CSTs, bilaterally. Comparison of metabolite measures between the groups were performed using ANCOVA with age as a covariate, and associations between the metabolite and clinical measures were evaluated using Pearson's product moment partial correlation. A p-value of < 0.05 was considered significant.

Results: In the whole-CSTs of ALS: Highly significant changes (p: 0.005 to < < 0.0002) found for the metabolites in both the sides were: 1) 7.5% and 4.6% decrease for NAA in the left and right, respectively; 2) 19.6% and 15.1% increase for Cho/NAA for the left and right, respectively. The NAA and Cho/NAA showed significant correlations with the rate of finger taps (NAA-left: r = 0.58, p = 0.0001; NAA-right: r = 0.40, p = 0.01; Cho/NAA-left: r = -0.546, p = 0.0001; Cho/NAA-right: r = -0.470, p = 0.003).

In the lobes of ALS: In the WM, NAA decreased by 6.0 % in frontal and 3.1% in parietal, and Cho/NAA increased by 10.1% in frontal, 5.4% in parietal, and 6.2% in occipital. For the GM, NAA decreased by 3.5% only in right-frontal, and Cho/NAA increased by 6.75% in frontal. The rate of finger taps correlated with NAA in the frontal WM (r = 0.348, p = 0.03) and GM (r = 0.349, p = 0.03) and with Cho/NAA in the WM (r = -0.329, p = 0.04).

Discussion and conclusions: The observed significant metabolite changes from the frontal, parietal and occipital lobes indicate that the ALS pathology is not anatomically localized only to the MC and CSTs but is more widespread. Significant correlations between NAA in the frontal lobe and CSTs and the rate of finger tapping support use of NAA as a biomarker of UMN dysfunction. Whole-brain MRSI and its unbiased metabolite analyses are shown to be well-suited for evaluating ALS pathology.

Funding: R01-NS060874 and R01-EB000822.

Reference

• V. Govind, et al. J Neurotrauma 2010;27:483–96.
   PubMed Web of Science ®Google Scholar

P79 CORTICAL IMAGING: AN UPPER MOTOR NEURON MARKER FOR AMYOTROPHIC LATERAL SCLEROSIS

VERSTRAETE E, VELDINK JH, VAN DEN HEUVEL MP, VAN DEN BERG LH

University Medical Center Utrecht, Utrecht, The Netherlands

Email address for correspondence: [email protected]

Keywords: MRI, neuroimaging, cortical thickness

Background: Upper motor neuron involvement for the diagnosis of amyotrophic lateral sclerosis (ALS) can be difficult to establish. An objective upper motor neuron marker would facilitate diagnosis. MR imaging is a widely available tool to study the brain anatomy. Computational techniques allow for detecting subtle degenerative effects on the brain.

Objectives: To study the degenerative effects of ALS on the morphology of the whole cortical mantle with an unbiased surface based approach.

Methods: We performed surface based cortical morphology analyses on structural, 3T MRI data of 45 patients with ALS and 25 age and gender matched healthy controls. Cortical morphology analyses consisted of measuring cortical thickness, surface area and volume. The effects of disease progression were examined by longitudinal measures in a subset of patients and correlating cortical measures with progression rate.

Results: Cortical morphology analyses revealed specific thinning in the precentral gyrus, being the primary motor cortex, in patients with ALS compared to controls (p = 6.3×10⁻⁸). Individual measures of the cortical thickness in the precentral gyrus discriminated patients with ALS from the healthy controls with a specificity of 82% and a sensitivity of 84%. Regarding the three morphology measures (thickness, surface area and volume), cortical thickness was found superior in detecting the degenerative effects of ALS. Relative cortical thinning in temporal regions was related with a faster disease progression (p = 3.3×10⁻⁴).

Conclusions: Cortical thinning of the primary motor cortex is a sensitive marker for upper motor neuron degeneration, proposing cortical imaging as a promising diagnostic marker for ALS. Progression of disease did not reveal additional cortical thinning in the primary motor cortex, suggesting these effects are potentially present before clinical diagnosis. Relative cortical thinning in temporal regions, however, was associated with a more progressive disease course.

P80 INCREASED 18F-FDG UPTAKE IN SUB-CORTICAL STRUCTURES OF ALS PATIENTS

CHIO A^1,2, PAGANI MM³, CALVO A¹, CANOSA A¹, GALLO S¹, MOGLIA C¹, VALENTINI MC⁴, MORBELLI S⁵, MONTUSCHI A¹, SALMASO D³, RODRIGUEZ G⁵, NOBILI F⁵, MANCINI M⁶, CISTARO A⁶

¹Department of Neuroscience, University of Torino, Torino, Italy, ²Neuroscience Institute of Torino, Torino, Italy, ³Institute of Cognitive Sciences and Technologies, Roma, Italy, ⁴Department of Neuroradiology, CTO Hospital, Torino, Italy, ⁵Clinical Neurophysiology Unit, Department of Neurosciences, Ophthalmology and Genetics University of Geno, Genova, Italy, ⁶Positron Emission Tomography Center IRMET, Torino, Italy

Email address for correspondence: [email protected]

Keywords: 18-F FDG PET/TC, sub-cortical areas, metabolic increase

Background: Recent data suggest that ALS is a multisystem disorder, characterized by the involvement of various central nervous system areas besides the motor cortex.

Aim: To assess PET/CT brain metabolic changes in a series of consecutive ALS patients.

Methods: ¹⁸F-FDG PET/CT scans of 32 ALS patients (13 bulbar, 19 spinal) were compared to those from 22 healthy controls. Differences were analyzed by statistical parametric mapping (SPM2), controlled by age and gender. A p < 0.05 threshold was used for SPM t-maps, corrected for multiple comparisons with the False Discovery Rate (FDR) option at voxel level and p < 0.001 corrected for multiple comparison at cluster level.

Results: Compared to controls, ALS patients showed a significant lower FDG uptake in the right ® and left (L) premotorcortex (Brodmann Area (BA) 6), R and L lingual gyrus (BA 18), R primary visual cortex (BA 17), R fusiform gyrus (BA 18) and L pre-central gyrus (BA 8, 9). The reverse comparison resulted in higher FDG uptake in ALS patients in the R and L amygdala, R and L pons and midbrain, R and L cerebellar tonsil, R lateral globus pallidus. ALS patients showed also a highly significant increase in FDG uptake in the brainstem (subthalamic nucleus, substantia nigra, and red nucleus). Bulbar onset patients showed high FDG uptake in R and L pons, and low uptake in almost all fronto-temporal regions; in spinal patients, FDG uptake was higher in R and L midbrain (BA 21, 28), and was reduced in R and L lingual gyrus (BA 18) and R fusiform gyrus (BA 19).

Conclusions: This PET/CT study revealed increased ¹⁸F-FDG uptake in large subcortical and brainstem regions in ALS. Significant differences between bulbar and spinal onset cases were also found. We hypothesize that the increase of glucose metabolism may be related to a reactive astrocytosis and/or microgliosis.

P81 DIFFUSION TENSOR IMAGING (DTI) MEASURES IN THE CORTICOSPINAL TRACTS AND THEIR CORRELATIONS WITH UPPER MOTOR NEURON FUNCTION MEASURES IN PATIENTS WITH ALS

SHARMA K, GOVIND V, SHERIFF S, SAIGAL G

University of Miami Miller School of Medicine, Miami, FL, USA

Email address for correspondence: [email protected]

Keywords: finger-and foot tap rate, diffusion tensor imaging, corticospinal tract

Background: Regions-of-interest (ROI) analyses of diffusion tensor imaging data have shown a reduction in fractional anisotropy (FA) along the corticospinal tracts (CST). However, its correlation with motor disability and clinical measures of upper motor neuron (UMN) function is still not clear or well established to be incorporated into diagnostic criteria.

Objective: To evaluate association between a DTI measure, fractional anisotropy (FA) and clinical measures of UMN function.

Design and methods: Fifty-one ALS patients (49 definite, 2 probable; 43 limb onset; 53 ± 9 years, disease duration: 19 ± 10 months) and 49 age-matched controls (51 ± 9 years) had evaluation for upper motor neuron function (maximum finger-and foot tap rates, pa-pa-and la-la syllable repeat rates; UMN burden score based on definite UMN signs, such as Hoffmann sign, Babinski sign and pathologic muscle stretch reflexes or spasticity). The disease severity was assessed using ALSFRS-R and vital capacity. Subgroup of subjects, 15 definite ALS- patients and 14 age matched controls, had whole brain scan on a 3T scanner using a spin-echo based DTI sequence (TR/TE = 6400/87 ms, b = 0, 1000 s/mm2, 12 gradient directions, 34 slices, 3 mm thickness). DTI data were processed using DtiStudio (https://www.mristudio.org/), and 3 ROIs were drawn. These included the posterior limb of the internal capsule (ROI-1), the midbrain at the level of the cerebral peduncle (ROI-@) and at the mid-pons (ROI-3), and FA values were obtained.

Results: The important findings were: the mean-FA was lower in patients with ALS than controls in all ROIs, bilaterally, within the CSTs (right-ROI-1: 0.60 ± 0.06 vs 0.66 ± 0.03, p = 0.003; right-ROI-2:0. 65 ± 0.05 vs 0.71 ± 0.04, p = 0.001; right-ROI-3:0. 51 ± 0.06 vs 0.61 ± 0.06, p = 0.0001). The maximum finger tap rate (3.4 ± 0.9 vs 4.1 ± 0.5; p = 0.0001), foot tap rate (2.2 ± 1.1 vs 3.4 ± 0.4; p = 0.0001), pa pa syllable repeat rat (3.4 ± 0.0.9 vs 4.1 ± 0.5; p = 0.0001) and la la syllable repeat rate (3.3 ± 1.0 vs 4.1 ± 0.4; p = 0.0001) were significantly lower in ALS patients than controls. There were mild to strong positive correlations between FA at various levels along the CST and clinical measures of UMN function in all the subjects (r = 0.4 – 0.8) except the UMN burden score, which correlated poorly with FA from all the ROIs (r = 0.2-0.4). ALSFRS-R poorly correlated with FA from all the ROIs (r = 0.2-0.5).

Conclusions and relevance: The reduced-FA reflects the functional abnormality of CST and its positive correlation with UMN function measures suggests that it can be used to assess UMN involvement in ALS patients objectively, and it may therefore contribute to earlier diagnosis of the disease.

This study was supported by the Stanley Glaser Foundation and NIH Grant # RO1 NS 060874

P82 WHOLE-BRAIN DTI PATTERN OF WHITE MATTER DAMAGE IN AMYOTROPHIC LATERAL SCLEROSIS: FURTHER EVIDENCE OF A MULTISYSTEM DISORDER

TROJSI F^1,2, ESPOSITO F^2,3, CIRILLO M^1,2, CAIAZZO G², SAGNELLI A¹, PICCIRILLO G¹, TEDESCHI G^1,2, MONSURRÒ MR^1,2

¹Department of Neurological Sciences, Second University of Naples, Naples, Italy, ²MRI Research Center SUN-FISM - Neurological Institute for Diagnosis and Care “Hermitage Capodimonte”, Naples, Italy, ³Department of Neuroscience, University of Naples “Federico II”, Naples, Italy

Email address for correspondence: [email protected]

Keywords: diffusion tensor imaging (DTI), tract based spatial statistics (TBSS), axonal injury

Background: Even if nowadays amyotrophic lateral sclerosis (ALS) is not anymore considered a basically motor disease, the actual spread of the neurodegenerative process throughout the central nervous system is not fully understood. In this work, we performed whole-brain tract-based spatial statistics (TBSS) and volume-of-interest (VOI) diffusion tensor magnetic resonance imaging (DTI-MRI) analyses to detect white matter (WM) patterns of microstructural abnormalities in ALS, and to correlate the DTI parameters with clinical indices of disability and pyramidal impairment.

Methods: Brain MRI and DTI were performed at 3 Tesla on 19 ALS patients with upper motor neuron (UMN) score (a clinical measure of pyramidal impairment) ranging from 2 to 16 and disease duration (DD) ranging from 1 to 14 years, in comparison to 20 age- and sex-matched healthy volunteers. Group-level analyses of DTI data sets were carried out with the Functional MRI of the Brain (FMRIB) Software Library (FSL) software package.

Results: Compared with controls, ALS patients showed a significant decrease in the fractional anisotropy (FA) in the body of corpus callosum (CC) (p < 0.05, corrected). When analyzing the ALS group alone, at the VOI level we observed both FA decrease and radial diffusivity (RD) increase in the body of CC that were significantly correlated with UMN score (p = 0.003 and p = 0.02). In addition, significant voxel-wise positive correlations between FA and ALS functional rating scale revised (ALSFRS-R) score (index of patients’ disability) were detected in the WM tracts underneath the left premotor cortex (p < 0.05, corrected).

Discussion and conclusions: The correlations between reduction of FA and increase of RD in the body of CC proportional to the UMN score suggest that the amount of WM degeneration in the CC is strictly related to ALS pyramidal impairment, and mainly determined by axonal loss within the motor fibers pathway. The correlation between FA and ALSFRS-R in the associative tracts underneath left premotor cortex might reflect the progressive spread of disease from motor towards extra-motor areas.

P83 VARIATE DIFFUSION TENSOR IMAGING ANALYSES FOR CHARACTERIZATION OF MOTOR NEURON DISEASES

MÜLLER HS-P, UNRATH A, LUDOLPH AC, KASSUBEK J

Department of Neurology, University of Ulm, Ulm, Germany

Email address for correspondence: [email protected]

Keywords: diffusion tensor imaging, tractwise fractional anisotropy statistics, variate analysis

Background: Pathophysiological processes in MND are related to the predominance of upper or lower motor neuron affection. Novel approaches based on computer-based techniques such as diffusion tensor imaging (DTI) are able to image disorders within the cerebral white matter (WM) in vivo. In a broad variety of studies, DTI has proven to be a valuable tool for analyzing these WM alterations.

Objectives: The objective of this study was to investigate patterns of CST and CC degeneration in different MND with involvement of the upper motor neuron by application of specific DTI analysis approaches.

Methods: Seventy-two patients with MND, i.e. amyotrophic lateral sclerosis (ALS, N = 20), primary lateral sclerosis (PLS, N = 20), hereditary spastic paraparesis (HSP), subdivided into pure HSP (pHSP), N = 20 and complicated HSP (cHSP), N = 12, were analyzed by application of variate DTI analyses in comparison with matched controls, in order to identify differences in FA. All analyses were performed by the Tensor Imaging and Fiber Tracking (TIFT) software (1). Analysis was performed in a variate fashion, i.e. voxelwise comparison of FA maps at the group level, volume of interest (VOI) analysis, as well as fiber tracking (FT) on group averaged data accompanied by tractwise fractional anisotropy statistics (TFAS) (2).

Results: The variate analyses of WM impairment demonstrated characteristic patterns of widespread alterations within the motor system for the various MNDs (3). A predominant affectation of the corticospinal tract (CST) and also WM changes within distinct areas of the corpus callosum (CC) were observed. In detail, the PLS group showed significant FA reductions in motor regions of the CC while the alterations of the pHSP group exceeded the motor region into the splenium and the changes of the ALS group to rostral parts of the corpus, respectively. Finally, the cHSP group showed significantly reduced FA within the whole CC and extensive areas of the subcortical WM.

Discussion and conclusions: In summary, variate DTI analysis was useful in order to define a distinct WM pathoanatomy in specific brain areas involving the CST and the motor segment (III) of the CC.

Reference

• Mueller HP, Unrath A, Ludolph AC, et al. Phys Med Biol 2007;52:N99–109.
   Google Scholar
• Mueller HP, Unrath A, Sperfeld AD, et al. Biomed Eng Online 2007;6:42.
   PubMed Web of Science ®Google Scholar
• Unrath A, Mueller HP, Riecker A, et al. Human Brain Mapping 2010;31:1727–40.
   PubMed Web of Science ®Google Scholar

P84 VOXEL-BASED RELAXOMETRY OF T2 MAPS IN THE BRAIN OF AMYOTROPHIC LATERAL SCLERORSIS

CHEN Z, MA L, CHEN Z, LING L

PLA General Hospital, Beijing, China

Email address for correspondence: [email protected]

Keywords: voxel-based relaxometry, magnetic resonance imaging, T2 relaxometry

Background: Relaxation describes the processes that spin magnetization prepared in a non-equilibrium state returns to the equilibrium distribution in magnetic resonance imaging. Relaxometry refers to the measurement of relaxation magnetic resonance imaging. Voxel-based relaxation is more sensitive to changes of tissue properties, i.e. differences in relaxation rates for instance increased iron deposition.

Objectives: To evaluate the characteristics of T2 relaxometry in the brain of amyotrophic lateral sclerorsis (ALS) patients using voxel-based relaxometry (VBR).

Methods: 32 definite or probable ALS patients based on E1 Escorial standards and 32 healthy controls were recruited and underwent a neuropsychological evaluation. The T2 mapping sequence data was collected on a GE Medical 3.0T MRI system. ANCOVA was applied with age as a covariate because of an exact sex match. A statistical threshold of P < 0.001 (uncorrected, t > 3.23) and more than continuous 20 voxels determined significance. The correlation analysis was applied between the mean intensity of active clusters and the progression scores by SPSS 15.0 and the curve fitting toolbox of matlab (version 7.6.0.324).

Results: VBR identified the T2 values increase in the right superior temporal gyrus white matter and inferior parietal lobule gray matter, the left superior frontal gyrus white matter and inferior frontal gyrus gray matter (P < 0.05). No areas of statistically significant decrease was displayed.

Discussion and conclusions: Voxel-based relaxometry demonstatred the good assessment of T2 value differences through the brain in ALS, and it could be regarded as an alternative means of relaxometry data analysis to evaluate the iron deposition in the brain of ALS.

P85 IS THERE A DISTINCT CEREBRAL PATHOLOGY IN PLS? EXPERIENCE WITH A NOVEL MYELIN-FOCUSED MRI RELAXOMETRY SEQUENCE

KOLIND S, TALBOT K, TURNER M

Oxford University, Oxford, UK

Email address for correspondence: [email protected]

Keywords: biomarker, myelin, inflammation

Background: Cerebral Wallerian demyelination and related neuroinflammatory processes have not been systematically studied in motor neuron diseases. Multi-component Driven Equilibrium Single Pulse Observation of T1 and T2 (mcDESPOT) is a novel MRI relaxometry technique allowing assessment of the myelin water fraction (MWF - a measure related to myelin content), and estimation of the myelin intra/extra-cellular (IE) water T2 properties linked to the presence of inflammatory processes.

Objectives: To use mcDESPOT as a marker of demyelination and inflammatory processes in heterogeneous cases of ALS, a group of definite PLS patients, and a group of healthy controls.

Methods: Data were acquired at 3T. Eighteen ALS patients of variable upper motor neuron (UMN) involvement clinically, 7 definite PLS patients and 9 healthy controls were included in the study. Image processing was performed to derive voxel-wise MWF and IE-water T2 maps. A WM skeleton was created from the MWF maps using tract-based spatial statistics (TBSS). Voxel-wise non-parametric testing was performed on the skeletonised MWF and IE-water T2 data to determine areas of group difference (ALS-control; PLS-control; ALS-PLS).

Results: Reduced MWF and increased IE-water T2 was found in both patient groups. MWF was able to discriminate the PLS group from controls, where widespread and symmetrical regions of decreased MWF were noted throughout the WM skeleton (average 4.8% decrease, and 2.6% across the entire WM skeleton; both p < 0.05), with areas of increased IE-water T2 in parts of the descending corticospinal tract (CST) and also corpus callosum (mean 5.2% increase in areas denoted in figure 1; 1.4% across the entire WM skeleton; both p < 0.05). In the ALS group, the MWF was not significantly decreased compared to controls (mean 0.7% lower across the WM skeleton), including in a sub-group analysis of just those with very high UMN scores, and those with a disease duration of more than 5 years. There were regions (largely motor) where MWF was significantly lower for PLS patients compared to ALS (regional mean 4.8% lower in PLS, 1.8% across the WM skeleton), with more widespread increased IE-water T2 (mean 3.7% increase in significant areas, 0.8% for the entire WM skeleton). There was no correlation between imaging parameters and disability for either group.

Discussion: mcDESPOT revealed a distinct pattern of reduced MWF in PLS compared to ALS patients, that appeared to transcend both UMN burden of disease clinically and disease duration, and suggests widespread demyelination. There was also indirect evidence of inflammatory activity from the increased regional IE-water T2 found in both patient groups, which may have implications for therapeutic intervention as well as understanding prognostic heterogeneity.

Conclusions: mcDESPOT has the potential to reveal novel aspects of in vivo pathology non-invasively, and adds to the growing biomarker potential of advanced MRI applied to motor neuron diseases.

P86 STRUCTURAL AND FUNCTIONAL CONNECTIVITY APPEAR INVERSELY RELATED WITHIN THE ALS CEREBRAL NETWORK

DOUAUD G, FILIPPINI N, TALBOT K, TURNER M

Oxford University, Oxford, UK

Email address for correspondence: [email protected]

Keywords: network, cerebral connectivity, biomarker

Background: ALS as a system failure is a concept based on two decades of research confirming a consistent extra-motor CNS pathology, and supported by recent functional MRI studies revealing altered resting-state networks (1). There is a need to understand the functional alterations associated with structural changes detected using advanced MRI techniques such as diffusion tensor imaging (DTI, for investigating white matter tract integrity) and voxel-based morphometry (for detecting grey matter volume changes).

Objectives: To characterize the functional connectivity changes within the ALS-specific cerebral network.

Methods: This was a cross-sectional multimodal MRI study in a heterogeneous group of ALS patients (n = 26) recruited from the ongoing longitudinal Oxford Study for Biomarkers in MND. Patients were compared to age and gender-matched healthy control subjects (n = 15). All subjects underwent high-field (3T) T1-weighted (1 mm isotropic), DTI (2 mm isotropic, 60 directions,), and resting-state functional MRI (3 mm isotropic). Tract-based spatial statistics (TBSS) on the DTI data characterized the white matter regions significantly altered in the ALS group. Tractography was used to identify grey matter regions linked to this impaired white matter network. A dual-regression analysis of the whole-brain resting-state functional MRI data was performed using this ALS-specific tractography-derived structural network to determine any modified functional connectivity in relation to abnormal structural connectivity.

Results: TBSS confirmed bilateral corticospinal and corpus callosum white matter tract involvement (2). Associated grey matter regions included motor, premotor and supplementary motor cortices, pars opercularis and motor-related thalamic nuclei. A spatial pattern of increased functional connectivity matched the pattern of decreased structural connectivity in all grey matter motor-related areas. Patients with longer disease duration showed relatively higher white matter tract integrity but lower functional connectivity, with values closer to those of the healthy controls in all instances. A composite measure of grey matter volume, structural connectivity functional connectivity information was ∼95% accurate for the discrimination of ALS patients.

Discussion: A combined structural and functional approach identified dichotomous processes characterizing the ALS cerebral network failure, involving increased functional connectivity within regions of decreased structural connectivity. Whilst this combination might reflect a compensatory response, an upstream dysregulation driving pathogenesis is plausible, possibly arising through loss of inhibitory neuronal influences (an ‘interneuronopathy’) given the similarity in regional loss of [11C]-flumazenil binding noted in a PET study (3).

Conclusions: Integrating multimodal MRI reveals the ALS cerebral signature in vivo and may generate surrogate markers of disease activity. Longitudinal, combined structural and functional studies, including the study of pre-symptomatic carriers of pathogenic ALS-related gene mutations, may ultimately allow the development of a Braak-like in vivo model of ALS pathogenesis.

Reference

• Mohammadi B, et al. Exp Neurol 2009;PM:19416664.
   Google Scholar
• Filippini N, et al. Neurology 2010;PM:21041787.
   Google Scholar
• Turner MR, et al. Brain 2005;PM:15843422.
   Google Scholar

P87 DIRECT EVIDENCE OF INTRA- AND INTER-HEMISPHERIC CORTICOMOTOR NETWORK DEGENERATION IN ALS: AN AUTOMATED MRI STRUCTURAL CONNECTIVITY STUDY

ROSE S¹, PANNEK K¹, BAUMANN F^1,2, HUTCHINSON N², COULTHARD A^1,3, MCCOMBE P^1,2, HENDERSON R^1,2

¹UQCCR, University of Queensland, Brisbane, Australia, ²Dpt Neurology, Royal Brisbane and Women's Hospital, Brisbane, Australia, ³Medical Imaging, Royal Brisbane and Women's Hospital, Brisbane, Australia

Email address for correspondence: [email protected]

Keywords: MRI, tractography, connectivity

Background: There is a growing interest in developing non-invasive neuroimaging biomarkers that can be used to improve our understanding of the pathogeneses of ALS and provide improved measures of disease progress. We have developed a fully automated technique for measuring the structural connectivity of corticomotor white matter (WM) pathways, based on diffusion MRI tractography in conjunction with cortical parcellation of high resolution structural images, that is optimised for resolving fibres within complex WM architecture (1).

Objectives: Our hypothesis is that quantitative intra- and inter-hemispheric structural connectivity measures based on assessing the Fractional Anisotropy (FA) for each corticomotor pathway will provide new insight into corticomotoneuron involvement in ALS. The reproducibility of the corticomotor connections were also assessed in control participants.

Methods: Structural MRI and HARDI diffusion data (64 diffusion encoding directions, b = 3000 mm/s²) were acquired from 15 ALS patients with mixed upper and lower motor neuron signs and 18 controls using a 3T scanner. Connectomes were generated using our automated pipeline (1). Each element within the connectivity matrix (i.e. cortical connection) was encoded with the mean FA value along that trajectory. Statistically significant differences in corticomotor connectivity between the ALS and control group were detected by applying a nonparametric Mann-Whitney test (applying a FDR of 10%). Eight control subjects were scanned twice over a period of 6 months to assess reproducibility.

Results: The reproducibility of the corticomotor connections was high (within 5%). In ALS patients there was significant loss in connectivity within a number of corticomotor pathways. With regards to altered intrahemispheric connectivity, there was a significant reduction in FA within the corticospinal connections of the brainstem (bs) with left precentral gyrus (lh.preCG), bs and right precentral gyrus (rh.preCG), bs and right postcentral gyrus (rh.postCG). Other intrahemispheric pathways significantly involved with ALS pathology were connections between the lh.preCG and left posterior cingulate (lh.postC), the rh.postCG and right paracentral gyrus (rh.paraCG), and the rh.postCG and right posterior cingulate (rh.postC). Novel results were also found for a number of interhemispheric pathways including the lh.preCG and rh.preCG, the lh.preCG and rh.postCG, and lh.preCG and right superior frontal gyrus (rh.supFG).

Discussion: The reduced connectivity associated with the precentral and postcentral gyri is consistent with known pathology and highlights the spread of neuropathology along multiple corticomotor pathways in ALS and supports a mechanism involving anterograde corticomotor neuron degeneration (2) in patients with mixed upper and lower motor neuron signs.

Reference

• Pannek K, Mathias J, Bigler ED, et al. Neuroimage 2010; 53(3):1044–53.
   Google Scholar
• Eisen A, Kim S, Pant B Muscle Nerve 1992;15:219–224.
   PubMed Web of Science ®Google Scholar

P88 CONTRIBUTION OF SPINAL MRI IN AMYOTROPHIC LATERAL SCLEROSIS

PRADAT P-F¹, EL MENDILI M-M², COHEN-ADAD J^2,3, MORIZOT-KOUTLIDIS R⁴, LEHERICY S⁵, BLANCHO S⁵, ROSSIGNOL S⁶, BENALI H²

¹Neurological Department, AP-HP, Pitié-Salpêtrière Hospital, Paris, France, ²UMR-678, INSERM-UPMC, Pitié-Salpêtrière Hospital, Paris, France, ³A.A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA, ⁴Department of Neurophysiology, AP-HP, Pitié-Salpêtrière Hospital, Paris, France, ⁵Centre for Neuroimaging Research (CENIR), Centre de Recherche de l'Institut du Cerveau et de la Moelle Epiniere, Pitié-Salpêtrière Hospital, Paris, France, ⁶Institut pour la Recherche sur la Moelle Epinière et l'Encéphale, Paris, France, ⁷GRSNC, Faculty of Medicine, Université de Montréal, Montreal, Canada

Email address for correspondence: [email protected]

Keywords: MRI, sensory pathways, spinal cord

Background: Amyotrophic lateral sclerosis (ALS) is a fatal disease characterized by upper (i.e. corticospinal tract, CST) and lower motor neuron degeneration. However, histopathological studies have shown involvement of spinal sensory pathways, supported by more recent in vivo imaging of the brain. These findings bring new insights to the pathogenesis of ALS. Characterizing in vivo spinal lesions in ALS is crucial to explore the anatomical structures affected by ALS. Novel MRI techniques such as diffusion-weighted (DW) and magnetization transfer (MT) imaging provide sensitive markers of white matter pathology.

Objectives: To combine DW and MT imaging of the cervical spinal cord in ALS patients and assess the presence of CST and dorsal column degeneration.

Methods: Patients with ALS (N = 29, Mean age = 53 ± 10 years, Median disease duration = 1.4 years) and age-matched controls (N = 21) were recruited. Patients were clinically assessed and scored on the ALS Functional Rating Scale (ALSFRS-R) and motor evoked potentials (MEP) were obtained using Transcranial Magnetic Stimulation (TMS). Subjects were scanned at 3T with DW and MT imaging. Manual ROIs were drawn in the spinal cord to isolate lateral (containing the CST) and dorsal (containing sensory afferents) segments. DTI and MT metrics were quantified in both ROIs and compared between patients and controls, then correlations with clinical scores and TMS measures were computed.

Discussion and conclusions: In the lateral region (CST), significant differences were detected between patients and controls for FA (p < 0.001), radial diffusivity (p < 0.001) and MTR (p < 0.01). No significant difference was detected for axial diffusivity (p = 0.15) and MD (p = 0.05). More interestingly, significant differences were detected in the dorsal column for FA (p < 0.001), radial diffusivity (p < 0.01) and MTR (p = 0.02). Pearson's coefficient showed significant correlations between FA in the lateral (CST) region and ALSFRS-R (p = 0.04) and TMS motor threshold (p = 0.02).

P89 ALS AND BODY MOVEMENTS – COMPENSATION IN HIGHER ORDER PROCESSING AREAS

HEIMRATH J¹, LULÉ D^1,2, LANG D^1,2, BIRBAUMER N^2,3, LUDOLPH AC¹

¹University of Ulm, Ulm, Germany, ²Eberhard Karls Universität Tübingen, Tübingen, Germany, ³Ospedale San Camillo, IRCCS, Venice, Italy

Email address for correspondence: [email protected]

Keywords: cortical plasticity, movement imagery, fMRI

Background: Execution, imagery, and perception of movements share similar neuronal substrates. A hallmark of ALS is progressive neuronal loss in primary motor areas. In previous studies, ALS patients showed more activity in higher order motor networks than healthy controls during execution and imagery of simple hand movements suggesting compensatory processes.

Objectives: In the present study we investigated if such compensatory neuronal processes can be seen for imagery and perception of more complex body movements of everyday life.

Methods: In this study, 7 ALS patients and 14 healthy controls were investigated using functional magnetic resonance imaging (fMRI). Thirteen movements were used for kinesthetic movement imagery and visual perception: four isolated movements (e. g., writing), four body related movements (e. g., tooth brushing), four movements which can be performed also in later stage ALS (e. g., blinking), and one control movement (rolling ball). Before fMRI measurement, subjects practised the kinesthetic imagery of each body movement and the visual imagery of the rolling ball; surface EMG electrode recording was performed to monitor possible subtle muscular activity during imagery. Additionally, motor impairment of the patients was assessed using the ALS functional rating scale revised (ALS-FRS-R). Several cognitive and psychiatric measurements were performed with all subjects.

Results: Preliminary data analysis suggests similar activation in ALS patients and controls during movement perception mainly in extrastriate visual areas (e. g., V2, V3), areas involved in higher order visual processing (e. g., MT), and premotor areas (BA 6). Patients showed more cortical activity during movement perception than healthy controls in areas for higher order movement representation (e. g., BA 40), whereas controls showed higher activity mainly in the right premotor cortex.

During movement imagery both groups similarly activated premotor areas, a network known as mirror neuron system (BA 44, BA 45), and higher order visual areas (e. g., BA7). ALS patients showed more activity in the premotor cortex than controls during this task. In contrast, healthy controls showed higher activity than patients in subcortical (e. g., putamen) and cortical (e. g., hippocampus) structures related to motor memory. The more advanced the disease, the stronger the activity in areas of higher order movement representation (e. g. BA 40).

Discussion and conclusions: The findings suggest compensatory processes in ALS patients during visual perception as well as imagery of body movements of everyday life. ALS patients’ increased activity in higher order visual areas together with their reduced activity in premotor areas during movement perception might indicate cross-modal compensation to overcome functional loss in motor networks. Progressive impairment of movement execution in ALS might lead to a fading motor memory of these movements. Increased activity in premotor areas and in areas of higher order movement representation might compensate for this in imagery tasks.

P90 SACCADIC EYE MOVEMENTS IN MOTOR NEURON DISEASE AND FRONTOTEMPORAL DEMENTIA

BURRELL J^1,2, HORNBERGER M^1,3, CARPENTER R⁴, KIERNAN M^1,2, HODGES J^1,3

¹Neuroscience Research Australia, Sydney, Australia, ²Prince of Wales Clinical School, Prince of Wales Hospital, Sydney, Australia, ³University of New South Wales, Sydney, Australia, ⁴Department of Physiology, Development and Neuroscience, University of Cambridge, Cambridge, UK

Email address for correspondence: [email protected]

Keywords: saccadometry, executive dysfunction, frontal dysfunction

Background: Motor neuron disease (MND) and Frontotemporal dementia (FTD) share clinical, pathological and genetic characteristics and saccadic eye movements may be abnormal in MND.

Objective: To characterise saccades in MND compared to FTD and controls using the Linear Approach to Threshold with Ergodic Rate (LATER) model.

Methods: Saccadometry was performed on MND and FTD patients, using a portable saccadometer, and results were compared to matched control subjects. Median reflexive saccadic latency, velocity, and LATER parameters (μ, ς, ςE) were computed, as was performance on an anti-saccade task. Parameters were correlated with cortical atrophy using VBM analysis in FTD patients.

Results: 70 subjects (25 MND; 16 limb-MND, 9 bulbar-MND, 22 FTD and 23 controls) were studied. Median saccadic latency was normal in MND, but prolonged in FTD (MND 201.1 +/− 31.3ms; FTD 229.6 +/− 41.8ms; controls 193.7 +/− 38.7ms, P < 0.05). Peak saccadic velocity did not differ in either patient group compared to controls. A measure of decision-making speed (μ) was normal in MND, but reduced in FTD (MND 5.1 +/− 0.9; FTD 4.5 +/− 0.9; controls 5.4 +/− 1.3, P < 0.05). ςE was increased in MND and FTD compared to controls, indicating an increased proportion of early saccades in both disease groups (MND 4.0 +/− 2.8; FTD 4.2 +/− 3.0; controls 1.1 +/− 2.4, P < 0.05). MND patients performed normally on the antisaccade task, but FTD patients performed poorly (MND 67.9 +/− 12.4%; FTD 56.5 +/− 9.9%; controls 66.7 +/− 13.2%, P < 0.05). μ and ςE correlated with atrophy of the left frontal eye field in FTD patients.

Discussion: Both MND and FTD patients had an increased proportion of early saccades, indicated by an increased ςE, which may reflect impaired inhibition of early saccades by higher cortical structures. Decision-making speed, as indicated by μ, was normal in MND, but reduced in FTD. Reduced μ and increased ςE correlated with atrophy of the left frontal eye field.

Conclusion: The only abnormality detected in MND patients was an increased proportion of early saccades, which may reflect subtle frontal lobe dysfunction. In contrast, FTD patients had more widespread abnormalities including an increased proportion of early saccades, reduced median saccadic latency, and poor performance on the anti-saccade task.

P91 ASSESSMENT OF ALS DISEASE PROGRESSION WITH THE SIX-MINUTE WALK TEST

CHEAH B^1,2, BRODATY N¹, BOLAND R¹, VUCIC S^1,3, KIERNAN M^1,2

¹Neuroscience Research Australia, Randwick, New South Wales, Australia, ²Prince of Wales Hospital Multi-disciplinary MND Clinical Service, Randwick, New South Wales, Australia, ³Sydney Medical School, Westmead, New South Wales, Australia

Email address for correspondence: [email protected]

Keywords: longitudinal assessment, clinical trials, functional impairment

The six-minute walk test (6MWT) has been conventionally used to measure functional exercise capacity in patients with cardiorespiratory disease. Given that functional impairment is an important source of morbidity in amyotrophic lateral sclerosis (ALS), simple measures of exercise capacity need to be validated. The objective of the present study was to investigate the utility of the 6MWT in monitoring disease progression and functional exercise capacity in patients with ALS. 28 ALS patients (20 males; age, 58.3 (SD, 9.4); median disease duration, 24.6 months (inter-quartile range, 14.2 – 34.4 months)) were recruited from a specialised multi-disciplinary ALS clinic, and evaluated longitudinally for up to 45 weeks. 6MWT, Short Form-36 (SF-36; physical component score) and ALS Functional Rating Scale-revised (ALS FRS-r) were administered. Linear mixed effects models with random intercepts and slopes were used for data analysis. Distance walked over six minutes at baseline was reduced (mean, 65.2% predicted; SD, 17.4% predicted). Distance walked over six minutes declined linearly at a rate of 0.32% predicted per week (S.E., 0.09; P = 0.0005). Patients with bulbar-onset disease walked 14.7% predicted (S.E., 0.1% predicted) further than patients with limb-onset disease over the entire study (P = 0.07). ALS FRS-r and its gross motor subscore also underwent linear decline at rates of 0.2 units (S.E., 0.03 units; P < 0.0001) and 0.1 units (S.E, 0.02 units; P < 0.00001) per week respectively. The physical component score of SF-36 did not appear to diminish over time (P = 0.11). Rate of decline in 6MWT (% predicted) was correlated with rates of decline in ALS FRS-r (Pearson's r = 0.57; P = 0.002) and ALS FRS-r (gross motor subscore; Pearson's r = 0.43; P = 0.02). There was no correlation between 6MWT (% predicted) and the physical component score of SF-36. Given these findings, we propose that the 6MWT may be incorporated as a functional measure of exercise capacity and disease progression in ALS clinical trials.