21st MND Symposium

1 Department of Neurology, UMC Utrecht, Utrecht, Netherlands,2 Department of Medical Genetics and Rudolf Magnus Institute,Utrecht, Netherlands, 3 Department of Clinical Neuroscience,King’s College, London, United Kingdom, 4 Department of Clinical

Neuroscience, Ume å University, Ume å , Sweden, 5 Departmentof Neurology, University of Ulm, Ulm, Germany, 6 Department ofInternal Medicine, Erasmus University Medical Centre Rotterdam, Rotterdam, Netherlands, 7 Department of Epidemiology,Biostatistics and HTA, Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands, 8 Department of Neurology, Charité University Hospital, Humboldt-University, Berlin, Germany,9 Department of Neurology, Beaumont Hospital, Dublin, Ireland,10 Department of Neurology, assachusetts General Hospital, Charlestown, United States, 11 Department of Neurology, Jagiellonian

University, Krakow, Poland, 12 Experimental Neurology, University of Leuven, Leuven, Belgium, 13 Vesalius Research Centre, Flanders Institute for Biotechnology (VIB), Leuven, Belgium, 14 Department of Neurology, University Hospital Leuven, University

of Leuven, Leuven, Belgium, 15 Division of Molecular and Clinical Neurobiology, Department of Psychiatry, Ludwig-Maximilians-University, Munich, Germany, 16 Department for Neurodegenerative Diseases, Hertie Institute for Clinical Brain Research,University of T ü bingen, and German Center for Neurodegenerative Diseases, Tuebingen, Germany, 17 UCLA Center for Neurobehavioral Genetics, Los Angeles, United States

E-mail address for correspondence: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Keywords: genome-wide association study, CNV, rare variant

Background : Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease selectively affecting motor neurons in brain and spinal cord. Recent genome-wide association studies have identified several common variants (SNPs) that increase disease susceptibility. In contrast, rare copy-number variants (CNVs), which have been associated with several neuropsychiatric traits, have not been studied for ALS in well-powered study populations.

Objectives : To identify rare CNVs that increase disease susceptibilty in sporadic ALS.

Methods : We conducted a two-staged association study focused on rare CNVs, including over 19,000 individuals. We analysed genome-wide Illumina data and DNA samples using

stringent quality control criteria. We used PennCNV software for CNV detection and applied extra quality control fi lters to reduce the number of false-positive CNV calls. We tested each gene for association by comparing the number of CNVs affecting the gene in cases versus controls. Loci (genes) with a nominal Fisher Exact P value _ 0.01 and a frequency of 1% in controls were selected for follow-up after careful validation with TaqMan qPCR.

Results: In the discovery cohort of 1,875 ALS patients and 8,731 controls we identifi ed two loci that met our criteria for follow-up: the DPP6 locus (CNVs in 10 of 1,875 ALS patients versus 13 of 8,731 controls, OR _ 3.59, P _ 6.6 _ 10_ 3 ), and the 15q11.2 locus, containing NIPA1 , the gene causing hereditary spastic paraparesis (HSP) type 6 (CNVs in 8 of 1,875 ALS patients versus 3 of 8,731 controls, OR _ 12.46, P _ 9.3_10 _5 ). Validation experiments confi rmed the presence of all (n _ 25) tested CNVs. In our replication population of 2,559 ALS patients and 5,887 controls, the genes of interest again contained more CNVs in patients compared to controls, but did not meet our criteria for independent replication: DPP6 locus: 10 CNVs in 2,559 ALS patients versus 12 in 5,887 controls,

OR _ 1.92, P _ 0.097, pooled results: OR _ 2.64, P _ 1.4_10 _3 ; NIPA1 : 7 CNVs in 2,559 ALS patients versus 5 in 5,887 controls, OR _ 3.23, P _ 0.041, pooled results: OR _ 6.20,

P _ 2.2_10 _5 ).

Discussion: We identifi ed two genes that show suggestive evidence for association with ALS disease status. DPP6 has been suggested previously as a candidate gene for ALS,

while NIPA1 has not been associated with ALS before. Mutations in NIPA1 cause HSP type 6, a disease characterized by the selective death of (central) motor neurons, suggesting

an important role in motor neuron biology. Statistical power remains a problem in studies aimed at rare variants.

Conclusions : CNV analyses provide suggestive evidence for DPP6 and NIPA1 as candidate susceptibility genes for ALS.

P108 PRIORITIZATION OF ALS CANDIDATE

GENES THROUGH INTEGRATIVE PATHWAY

ANALYSIS

KENNA K 1 , HARDIMAN O 1,2 , BRADLEY D 1

1 Trinity College Dublin, Dublin, Ireland, 2 Beaumont Hospital,

Dublin, Ireland

E-mail address for correspondence: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Keywords: candidate , sequencing , pathway

Background: Mutations in 8 genes have been demonstrated

to co-segregate with familial ALS. Such variants cause disease through their infl uence on a single or multiple biological processes. Variation in other genes capable of infl uencing the same biological processes may also cause disease. For example genetic variation in APP, PSEN1 or PSEN2, can cause familial forms of Alzheimer’s disease. Continual advances in DNA sequencing based technologies provide the opportunity for signifi cantly larger scale

interrogation of the genomes of ALS patients for disease relevant variation. Our work is centred on elucidating the genetic basis for ALS in the Irish population by screening candidate

genes selected primarily by; 1) position of their encoded proteins within the human protein interaction network relative to proteins encoded by known ALS genes, and 2) similarity

of their proteins to ALS proteins in terms of sequence and annotation.

Methods: Candidate Selection: Experimentally verified human protein interactions were retrieved from online database and supplemented with interactions manually curated

from the literature to create a model of the human protein interactome. First order interaction partners of known ALS proteins and the second order interaction partners achieving

the best prioritization scores were selected as candidates. Predicted paralogs of ALS proteins were also included with second order interactors for prioritization. Prioritization technique: An in-house version of Google’s PageRank algorithm was written and used to score candidates

based on position within our model of the human protein interactome relative to that of known ALS proteins. The online tool ToppGene was used to score candidates based on similarity to ALS proteins in terms of numerous annotation features based on protein function, localization and expression. Candidates were also scored by sequence similarity to known ALS proteins using BLASTp. Scores were integrated to an overall score. A weighted profi le of genes potentially

relevant to ALS was constructed based on various lines of evidence such as occurrence within a known ALS linkage region etc and used to adjust fi nal prioritization.

Results: 283 fi rst and 4140 second order interaction partners of ALS proteins were identifi ed; 321 predicted paralogs were recovered based on an e-value threshold of 10; The coordinates

of the exons of 600 genes have been submitted to Agilent for RNA capture probe design to allow for subsequent sequencing.

Conclusion: Current technologies allow much higher throughput screening of candidate genes for disease relevant variation. Genes capable of infl uencing similar biological processes to known ALS genes are worthy candidates, and we have integrated multiple approaches to prioritize the most plausible of such candidates for sequencing in ALS patients.

148 Cognitive and Psychological Assessment and Support 21st International Symposium on ALS/MND

P172 A NEW METHOD OF CATEGORISING

NEUROPSYCHOLOGICAL DEFICITS IN PEOPLE

WITH ALS

JORDAN N 3 , PENDER N 1 , PHUKAN J 2 , ABRAHAMS S 4 ,

ELAMIN M 1 , BEDE P 1 , HARDIMAN O 1,2

1 Beaumont Hospital, Dublin, Ireland, 2 Trinity College Dublin, Dublin,

Ireland, 3 Royal College of Surgeons in Ireland, Dublin, Ireland,

4 Department of Psychology, Institute of Psychiatry, King’s

College London, London, United Kingdom

E-mail address for correspondence: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Keywords: consensus criteria, cognitive/behaviour impairment,

methodology

We present the first methodological analysis of the consensus criteria (1) in a population-based study of ALS, and we make our own recommendations for alternative analysis. The current paper is based on the initial dataset gathered by one of the authors as a part of a longitudinal Irish population-based study of ALS; the largest of its kind. Analysis of the results of an extensive neuropsychological battery (including tests of executive function, memory, language and visuospatial ability) from 51 patients with ALS, but with no co-morbid dementia, was carried out.

Results indicate that use of the Strong et al . (1) criteria identified 33% of patients with ALS as ALSci (ALS cognitive impairment; defined as below the fifth percentile on two or

more tests of executive function). Our recommended criteria identified 13.7% of patients with ALS as having executive dysfunction and a further 31.4% as having some other cognitive impairment (memory, multidomain, language or visuospatial). Therefore, based on our analysis, 45.1% of people with ALS had some form of cognitive impairment. Patients were categorized separately for behavioral impairment.

Results of an analysis of the Frontal Systems BehaviourScale (FrSBe) indicated that 23% of the 51 participants with ALS had behavioral impairment. In an attempt to avoid over-diagnosis of behavioural impairments, a classification of behavioural impairment is only given if the person scores more than 65 on the T-scores of 3 or more of the after subscales, unless the person’s before score on the same subscale was also over 65, thus demonstrating no change over time. Based on the results of the analysis, it appears that the Strong criteria may underestimate the number of people with cognitive impairment and the nature of impairment. The results of analysis using the Strong criteria suggests that 33% of bulbar and spinal onset participants combined were classifi ed as cognitively

impaired (ALSci), compared with 4% of the control participants. Our analysis demonstrated that in fact up to 45% of ALS participants may be impaired on one or more cognitive domains of interest. However, in our suggested analysis, a smaller percentage of 13.7% of the participants with ALS who have not been classifi ed with FTLD had executive dysfunction (5.9% executive dysfunction, 7.8% multidomain with executive dysfunction).

In addition, a large number had impairment in other areas, such as 7.8% with memory impairment, 11.8% with multidomain impairment without executive dysfunction and 9.8%

with language impairment. In addition, based on analysis using our suggested ethodology, 2.6% of controls met criteria for executive impairment, in contrast with the Strong criteria,

which found that 12% of controls had executive impairment, a number that seems high for a control population.

Reference:

1. Strong MJ, Grace GM, Freedman M, et al . AmyotrophicLateral Scler. Jun;10(3):131–46.

P173 COGNITIVE PROFILE OF ALS PATIENTS

WITH COMPARISON OF INCIDENT VERSUS

PREVALENT POPULATIONS: A DOMAIN-BASED

APPROACH

ELAMIN M 1,2 , PHUKAN J 1,2 , BEDE P 1,2 , JORDAN N 1 ,

PENDER N 1 , HARDIMAN O 1,2

1 Beaumont Hospital, Dublin, Ireland, 2Trinity College Institute of

Neuroscience, Dublin, Ireland

E-mail address for correspondence: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Keywords: cognitive, classifi cation, incident

Background: Estimates of cognitive impairment in patients with amyotrophic lateral sclerosis (ALS) vary widely. Studies published to date are small, clinic based and are heavily reliant on the prevalent ALS population. The literature is further

confounded by the currently used criteria (1) for cognitive impairment in ALS which focuses on executive dysfunction, when there is increasing evidence of involvement of other cognitive domains in ALS.

Objectives: 1) To describe the incidence of cognitive impairment in a population-based ALS cohort using new criteria that focus on single or multiple cognitive domain dysfunction; 2) To compare frequency of cognitive impairment in the incident

ALS patients to the prevalent ALS patients.

Methods: All cases of possible, probable or definite ALS in the Republic of Ireland have been invited to participate. Demographic, clinical and neuropsychological data are collected during home visits carried out at baseline and then repeated at six monthly intervals. Cognitive impairment in ALS patients is defined according to performance on neuropsychological tests covering five cognitive domains: attention/executive function, memory, language, visuospatial skills. Patients were then classified depending on

two factors: 1) the number of cognitive domains involved (none, single or multiple); 2) the presence or absence of executive dysfunction. The frequency of each cognitive syndrome

in the ALS population was compared to that in age, sex and education-matched healthy controls.

Results: To date, 167 patients have been recruited to the study. Mean age was 63.5 years, 60% were males and 33.6% of patients had bulbar-onset ALS. Incident cases comprised

83.3% of the cohort. Twenty two patients (13.6%) fulfilled the Neary criteria for

frontotemporal dementia (FTD). Compared to healthy controls, non-demented ALS patients had significantly higher frequency of single or multi-domain cognitive impairment

with executive dysfunction (29.9% versus 2.6%, P _ 0.0001). There was no significant difference between the 2 groups in frequency of single or multi-domain cognitive impairment without executive dysfunction (22.2% versus 16.5%,

P _ 0.311).

Incident ALS cases had a higher frequency of cognitive impairment compared to prevalent cases. The most striking difference was observed in the frequency of single domain

executive dysfunction but it did not reach statistical significance (11.7% versus 3.7%, P _ 0.39). The striking absence of executive dysfunction in the prevalent cases compared to the incident cases led to an investigation into the effect of executive dysfunction on patient

survival. The presence of executive dysfunction was found to be significantly associated with shorter survival in ALS patients (31 months versus 38 months, P _ 0.005) even after

correcting for multiple factors including age, bulbar onset and disease severity.

Discussion and conclusion: The incidence of FTD in a population based predominantly incident ALS cohort was 13.6%. Single and multi-domain executive dysfunction was

significantly more frequent in the ALS population compared to healthy controls. The incident population had a higher incidence of impairment in all cognitive domains, in particular executive function. The fact that executive dysfunction was

predominantly a feature of incident, rather than prevalent ALS patients, is likely to be secondary to its association with poorer prognosis.

Reference:

1. Strong MJ, Grace GM, Freedman M, et al . Amyotrophic

Lateral Scler. 2009;10(3):131-46.

P189 MAGNETIC RESONANCE IMAGING OF

HIRAYAMA DISEASE – A MIMIC CONDITION

OF MOTOR NEURON DISEASE

BEDE P 1,2 , HARDIMAN O 1,2

1 Trinity College, Dublin, Ireland, 2 Beaumont Hospital, Dublin,

Ireland

E-mail address for correspondence: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Keywords: Hirayama disease, MRI, monomelic amyotrophy

Background : Hirayama disease is a monomelic amyotrophy, a benign condition of the cervical spinal cord that can mimic motor neuron disease (MND). It frequently presents with unilateral or asymmetrical upper limb weakness and atrophy in young male patients. Electrophysiological studies are not specific to the condition. Anterior displacement of the posterior wall of the cervical dural canal is thought to be the underlying pathomechanism. The recognition and establishment of the diagnosis of Hirayama disease is challenging because of 158 Improving Diagnosis, Prognosis and Disease Progression 21st International Symposium on ALS/MND confounding clinical and electrophysiological features similar

to MND. Early diagnosis and prevention of flexion by a collar might prevent disease progression.

Method : Four patients with Motor Neuron Disease, 4 agematched healthy controls and 2 young patients with suspected Hirayama disease underwent cervical spinal cord imaging. A

3.0 Tesla MRI system was used to acquire 3D high resolution structural and diffusion tensor data of the cervical spinal cord in extension and fl exion. Multi-echo fast fi eld echo (mFFE) MRI sequence was used to enhance spinal grey matter – white matter contrast.

Results: The clinical suspicion of Hirayama disease was confirmed by demonstrating marked anterior shifting of the posterior wall of the cervical dural canal with cord fl attening in flexion. No similar changes have been observed on the healthy controls and motor neuron disease patients.

Conclusion: Standard non-flexion, neutrally positioned cervical spinal cord MRI might be non-specific to Hirayama disease. Dynamic myelography or flexion MRI studies are

required to confirm the diagnosis.

P193 SLOW SACCADES IN BULBAR-ONSET

MOTOR NEURONE DISEASE

DONAGHY C 1 , PINNOCK R 1 , FORBES R 1 , HARDIMAN

O 2 , PATTERSON V 1 , MCGIVERN CR 1 , GIBSON MJ 1

1 Royal Victoria Hospital, Belfast, United Kingdom, 2 Beaumont

Hospital, Dublin, Ireland

E-mail address for correspondence: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Keywords: saccades, ocular fi xation, PSP

Background: Current findings suggest that eye movement abnormalities in ALS relate to frontal lobe impairment. Numerous case reports, however, describe slow saccades and

supranuclear gaze palsies in patients with MND often associated with bulbar-onset disease. We performed a study of saccades and ocular fi xation in patients with MND to examine

for any differences between bulbar-onset, spinal-onset ALS and controls. We then compared the results for bulbar-onset ALS with a group of patients with PSP.

Methods: Forty-four ALS patients, 45 controls and 7 PSP patients were included. Refl exive horizontal saccades (latency and speed) and ocular fi xation (saccadic intrusion amplitude

and frequency) were examined using infra-red oculography. A Saccadic Intrusion (SI) Index was developed by the authors representing saccadic intrusion frequency and amplitude.

Results: Saccades were found to be slower in bulbar-onset compared to spinal-onset patients and controls (P _ 0.03, P _ 0.05). PSP patients had signifi cantly slower saccades comparedto controls (P _ 0.006). SI frequency was greater in PSP and bulbar-onset ALS patients compared to controls (P _ 0.006, 0.015). Similarly, SI Index was greater in PSP and

bulbar-onset ALS patients compared to controls (P _ 0.01, 0.002). Although not statistically signifi cant, PSP patients had increased SI freq, SI amplitude, SI Index and slower saccades

compared to bulbar-onset ALS patients.

Conclusions: This is the fi rst study to highlight the presence of slow saccades in bulbar-onset MND. It is likely that a spectrum exists, ranging from normal saccade speed to the presence of a gaze palsy. This theory is consistent with the similarities found in the eye movement profi le of PSP and bulbar-onset ALS patients. A longitudinal study would be of

great interest to examine the potential of eye movements as a biomarker for ALS disease progression. In addition, SI Index appears to be a sensitive measure of ocular fi xation abnormality that could be employed in further research

P199 NEURORADIOLOGICAL

CHARACTERISATION OF EXTRA MOTOR

CORTEX PATHOLOGY IN AMYOTROPHIC

LATERAL SCLEROSIS

BEDE P 1,2 , BOKDE A 1,2 , JORDAN N 2 , ELAMIN M 1,2 ,

FAGAN A 3 , HARDIMAN O 1,2

1 Trinity College, Dublin, Ireland, 2 Beaumont Hospital, Dublin,

Ireland, 3 St James Hospital, Dublin, Ireland

E-mail address for correspondence: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Keywords: MRI, cognition, frontotemporal dementia

Background : Amyotrophic Lateral Sclerosis is a complex multisystem disorder with non-motor manifestations: neuropsychological and behavioural defi cits. ALS with Frontotemporal Dementia (ALS-FTD) is increasingly recognized as a distinct phenotype of Motor Neuron Disease with distinguishing clinical, neuroradiological and neuropsychological features. Behavioural and cognitive deficits in ALS may precede

motor symptoms, are challenging to manage and have a significant impact on survival and the quality of life of carers and patients.

Aims : The aim of this study is to determine the extent of frontotemporal pathology in ALS patients with no overt cognitive impairment and to describe the Magnetic Resonance

Imaging characteristics of the ALS-FTD complex.

Methods : Thirty age matched healthy controls and 30 ‘ definite ’ ALS patients according to the El Escorial criteria have been recruited. Participants are assessed by a comprehensive

battery of neuropsychological tests and undergo 3 Tesla high resolution magnetic resonance neuroimaging. The neuropsychological battery includes the Boston naming test, Raven ` s Poster Communications Imaging, Electrophysiology and Markers of Disease Progression 163 standard progressive matrices, Logical Memory tests, the California verbal learning test, Verbal paired associates, the Rey Complex Figure test, the Brixton spatial anticipation test, Digit span, Verbal fl uency tests, Category Fluency, the Wechsler

test of adult reading, The colour word task – Stroop test, the Hospital Anxiety and Depression Scale and the Frontal Systems Behaviour Scale (FrSBe). The neuroimaging protocol includes: Voxel Based Morphometry (VBM) based on high resolution structural imaging data, Magnetic Resonance Spectroscopy (MRS), Diffusion Tensor Imaging (DTI) and Resting state Functional MRI (rfMRI). A routine Fluid-attenuated inversion recovery (FLAIR) sequence is used for the identifi cation of underlying cerebrovascular disease to exclude patients from the research study.

Results: Five ALS patients fulfi lled the Neary Criteria for Frontotemporal dementia. Fifteen ALS patients scored within normal range in all of the neuropsychological domains when compared to a pool of age matched healthy controls, and were categorized as ALS patients with no cognitive defi cits. Ten ALS patients demonstrated executive dysfunction on neuropsychological testing. Voxel based morphometry (VBM) analysis demonstrated considerable differences in grey matter volumes between ALS-FTD and ALS patients with no cognitive defi cits. Comparative statistical maps highlighted significant differences (P _ 0.05) in the left anterior temporal lobe, ventral medial frontal lobe and parahippocampal regions. ALS patients with no cognitive defi cits showed grey matter volume changes in the hippocampus, left dorsolateral prefrontal cortex, cerebellum, posterior cingulate and in the in right superior temporal gyrus when compared to healthy controls.

Conclusions: The results underscore the heterogeneity of extra motor involvement within the ALS spectrum. The study demonstrates the unique radiological attributes of the ASL-FTD complex. However, ALS patients with no cognitive or behavioral deficits on formal neuropsychological testing also show signs of extensive extra motor cortex atrophy on a group level.

CW229 PROPOSED CRITERIA FOR FAMILIAL AMYOTROPHIC LATERAL SCLEROSIS

BYRNE S1 ,2, WALSH C1, LYNCH C2, BEDE P1 ,2, ELAMIN M1 ,2, KENNA K1, MCLAUGHLIN R1, HARDIMAN O1 ,2

1Trinity College, Dublin, Ireland, 2Beaumont Hospital, Dublin, Ireland

Email address for correspondence: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Keywords: familial ALS, genetics, meta analysis

Introduction: Amyotrophic lateral sclerosis(ALS) is a devastating neurodegenerative disease. The rate of familial amyotrophic lateral sclerosis (FALS) is reported as 10%, but a systematic review and meta-analysis of the true population-based frequency of FALS undertaken by our group refutes this fact. Calculations show that chance plays a significant role in a second family member being affected. There is currently no consensus on the definition of FALS. We propose criteria for FALS which incorporate family history and genetic analysis.

Methods: Thirty-three papers were included in meta-analysis to calculate the pooled rate of FALS. The probability of a second family member being affected by ALS by chance was calculated. 9

Results: Pooling data from prospective population-based registry data revealed a rate of FALS of 5.1% (95% CI 4.1-6.1).

The lifetime risk of ALS in the UK is 1:411 and population statistics show the average kindred size to be 17. Thus a proband with 17 first- and second-degree relatives has a 4.1% chance of having another affected relative.

Discussion: There is currently no consensus on the definition of FALS. We have shown that the rate of FALS is lower than previously reported and that chance may play a significant part in two people being affected within a kindred. To increase the yield of genetic studies it is important to have a clear definition of FALS. This short report proposes criteria for FALS.

Conclusion: The true rate of FALS is 5%. A diagnosis of Familial ALS should be made with caution in kindreds with only two affected members as co-existence of disease within a family may be due to chance occurrence. We have proposed new criteria for FALS. The aim of these criteria are to increase the yield of genetic studies and to facilitate comparative interpretation of epidemiological and genetic FALS data.

References: Alonso A, Logroscino G, Jick S et al. Eur J Neurol 2009: 16: 745-751 Rabe M, Felbecker A, Waibel S et al. J Neurol 2010 DOI 10.1007/s00415-010-5512-9

CW230 THE USE OF A NOVEL GEO-CODING TECHNIQUE IN ALS EPIDEMIOLOGY

BYRNE S1 ,2, LYNCH C2, CULLEN CL3, JOHNSTON H3, HARDIMAN O1 ,2

1TCIN, Trinity College, Dublin, Ireland, 2Beaumont Hospital, Dublin, Ireland, 3Health Atlas Ireland, Stevens Hospital, Dublin, Ireland

Email address for correspondence: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

Keywords: epidemiology, etiology

Background: A small percentage of cases of ALS are accounted for by dominant gene mutations. The remaining sporadic form of the disease is thought to be due to a complex interplay between genetics and environment. Detailed population-based epidemiological analysis is necessary to identify environmental factors that may be implicated in disease pathogenesis.

There are conflicting reports as to whether there is a difference between the rate of ALS in urban and rural areas.

Objectives: 1) To provide every patient diagnosed in Ireland from 1995-2009 an x,y co-ordinate and map these to identify evidence of clusters; 2) To compare urban and rural areas; 3) To compare all mapped cases from 1995-2009 with environmental agency maps to identify any factors significantly associated with apparent disease clusters.

Methods: We have employed a population based epidemiological approach using geo-coding by Health Atlas Ireland (HAI). An x,y co-ordinate was given to every patient diagnosed with ALS between 1995-2009. Environmental Protection Agency (EPA) data is correlated with x,y patient co-ordinates to identify statistically significant clusters.

An algorithm is under development to quantitatively assess whether there is a significant difference between urban and rural areas and whether environmental factors are significantly associated with clusters.

Results: 1187 patient co-ordinates were mapped. Sporadic and familial cases were mapped to identify possible clusters. There is an increase in rate of ALS in north-western counties. An algorithm is under development to determine whether there is evidence of clustering in other regions of the country.

Discussion: This is the first study to engage mapping techniques to identify location of ALS cases in an entire country. An east-west gradient has been identified that reflects the underlying genetic structure of the Irish population. This study demonstrates the potential use of geo-coding methods used to interrogate the the geographic epidemiology of ALS in a defined population.