Risk factors for Amyotrophic lateral sclerosis · Amyotrophic lateral sclerosis (ALS) is said to be...

Risk factors for Amyotrophic lateral sclerosisLifestyle, environment and genetics

Meinie Seelen

2014226 Meinie Seelen_binnenwerk.indd 1 30-04-15 22:43

©2015 Meinie Seelen

All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronical or mechanical, including photocopy, recording, or any information storage or retrieval system, without permission in writing from the author. The copyright of the articles that have been published has been transferred to the respective journals.

ISBN: 978-90-393-6350-8Cover: Milou Bicker DesignDesign: wenz iD, Wendy SchoneveldPrinted by: Proefschrift All in One

The studies described in this thesis were performed at the Brain Center Rudolf Magnus, Department of Neurology, University Medical Center Utrecht, The Netherlands.

Funding of the studies described in this thesis was provided by the Netherlands ALS Foundation, Prinses Beatrix Fonds (PB 0703), VSB fonds, H Kersten and M Kersten (Kersten Foundation), J R van Dijk, the Adessium Foundation, the European Community's Health Seventh Framework Programme (FP7/2007-2013), ZonMW under the frame of E-Rare-2, the ERA-Net for Research on Rare Diseases, EU Joint Programme – Neurodegenerative Disease Research (JPND) project, the Netherlands Organisation for Health Research and Development (Vici scheme to LHvdB).

The publication of this thesis was financially supported by ChipSoft BV.


RISK FACTORS FOR AMYOTROPHIC LATERAL SCLEROSIS

LIFESTYLE, ENVIRONMENT AND GENETICS

Risicofactoren voor amyotrofische laterale scleroseLeefstijl, omgeving en genen

(met een samenvatting in het Nederlands)

PROEFSCHRIFT

ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van de rector magnificus, prof. dr. G.J. van der Zwaan,

ingevolge het besluit van het college voor promoties in het openbaar te verdedigen op dinsdag 16 juni 2015 des middags te 2.30 uur

door

Meinie Seelen

geboren op 12 januari 1986 te Arnhem


Promotoren: Prof. dr. L.H. van den Berg Prof. dr. J.H. Veldink

Copromotoren: Dr. M.A. van Es Dr. ir. R.C.H. Vermeulen



CONTENTS

CHAPTER 1 Introduction 9

PART I - LIFESTYLE

CHAPTER 2 Lifetime physical activity and the risk of ALS 17

CHAPTER 3 Prior medical conditions and the risk of ALS 31

CHAPTER 4 Presymptomatic BMI, dietary fat and alcohol consumption as independent risk factors for ALS

49

PART II - ENVIRONMENT

CHAPTER 5 Occupational exposure to diesel motor exhaust increases the risk of ALS

67

CHAPTER 6 Long-term exposure to air pollution is associated with an increased risk of ALS

85

CHAPTER 7 Residential exposure to extremely low frequency electromagnetic fields and the risk of ALS

101

PART III - GENETICS

CHAPTER 8 No mutations in hnRNPA1/A2B1 in Dutch patients with ALS, FTD and IBM

107

CHAPTER 9 Large scale genetic screening in sporadic ALS identifies modifiers in C9orf72 repeat carriers

115

CHAPTER 10 General discussion 137


ADDENDUM

Nederlandse samenvatting (Summary in Dutch)Dankwoord (Acknowledgements)List of publicationsAbout the author

150155158160



Introduction

CHAPTER 1


CHAPTER 1

10

Amyotrophic lateral sclerosis (ALS) is said to be one of the most devastating neurodegenerative disorders in adults. It is characterized by rapidly progressive motor neuron loss in the brain and spinal cord. Initial presentation of patients with ALS is usually muscle weakness of limbs or difficulties with speech. Eventually these symptoms progress to paralysis, swallowing difficulties and generally respiratory failure. Fifty percent of patients die within three years from symptom onset.1-3 Cognitive impairment is increasingly recognized as a feature and can be seen in up to 40% of ALS patients.4 To date, there is no treatment that can significantly slow or stop disease progression. Only one drug (Riluzole) is currently available, which offers just a modest prolongation of survival (by approximately 3 months) in patients with ALS.5 Each year about 400-500 people in the Netherlands are diagnosed with ALS.1 The lifetime risk of ALS is estimated at 1:400, with a median incidence of ALS of 2.8/100.000 person years. ALS can occur at any adult age, with a median age at onset of 63 years. Men are slightly more frequently affected than women, with a male-female ratio of ~1.5. Gender specific incidence rates show an evident decrease of the incidence after the age of 75 years, in contrast to other neurodegenerative disorders (Figure 1.1). This indicates that ALS is not a disease of ageing, such as Alzheimer’s and Parkinson’s disease, but that age is an important factor in its pathogenesis. The incidence rate, age and gender distribution in the Netherlands are similar to those described for other European countries.6, 7 Multiple pathophysiological mechanisms cumulate to cause motor neuron degeneration in ALS, such as glutamate excitotoxicity, oxidative stress, dysregulated RNA metabolism, neuroinflammation, mitochondrial dysfunction, protein aggregation and impaired axonal transport.8 However, most research has been performed in animal models or in specific genetic variants of familial ALS. The relative contribution of the different pathways for

0

2

4

6

8

10

12

14

15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 >85

Inci

denc

e (/1

00,0

00 p

erso

n-ye

ars)

Age group (years)

Men Women Total

Figure 1.1 Age and gender specific incidence rates of ALS in the Netherlands. Obtained from Huisman et al.1


11

Introduction

subtypes of ALS may vary and it is still unknown whether those mechanisms are applicable to all ALS patients.The majority of ALS patients appears to be sporadic, whereas only 5-10% of patients have a familial form of ALS, in which the index patient has a first- or second-degree family member with ALS.9 Familial ALS cases have enabled the identification of several genetic triggers, of which mutations in superoxidase dismutase 1 (SOD1), fused in sarcoma (FUS), transactive response DNA binding protein 43 (TARDBP) and chromosome 9 open reading frame 72 (C9orf72) are the most frequent.10

While such advances have contributed to our current understanding of the causes of most familial ALS cases, they only explain a small fraction of the far larger group of sporadic ALS cases. In sporadic ALS the etiology seems to be more complex, with multiple risk factors interacting to cause ALS: not only genetic factors, but also lifestyle and environmental factors. A twin study showed that the heritability component of sporadic ALS accounts for approximately 61%, with the remaining part explained by lifestyle and environmental factors.11 With this knowledge, the etiology of a large proportion of sporadic ALS cases remains still unexplained.

AimWithin this thesis, we aim to achieve a high level of evidence for genetic and epidemiological studies to identify risk factors that predispose to sporadic ALS. Elucidating risk factors provides new insight into pathogenic mechanisms of ALS and eventually may identify targets for novel therapeutics.

Prospective ALS study the NetherlandsIn 2006 the Prospective ALS study the Netherlands (PAN), a large nationwide population-based case-control study, was initiated by the ALS center the Netherlands.1 Within this study we continuously recruit all incident ALS patients in the Netherlands and subsequently include population based controls (matched on gender and age to the patients). Hereby, we avoid the risk of referral and selection bias. Moreover, previous studies showed that prevalent patients differ significantly from incident patients in age at onset, site of onset and survival,1 emphasizing the importance of inclusion of incident ALS patients. Patients with ALS had to meet the criteria for possible, probable (laboratory supported) or definite ALS according to the revised El Escorial criteria.12 These diagnostic criteria were developed and revised to enhance participation of patients in therapeutic trials and clinical and genetic research. All participants are registered in our national database, after which (a) medical records of patients are retrieved to obtain clinical characteristics, (b) all participants are asked to complete a detailed questionnaire on lifestyle and environmental factors, and (c) blood samples are collected for isolation of DNA.


CHAPTER 1

12

OUTLINE OF THE THESIS

Part I – LifestyleMany different lifestyle factors have been proposed in the pathogenesis of ALS, though smoking is the only exposure that has been consistently identified as risk factor.13 The evidence for other lifestyle factors (e.g. alcohol consumption, physical activity, prior trauma, dietary intake) is less consistent, in part due to low case numbers and poor study design. One perception of patients diagnosed with ALS is that they are “fitter” before clinical onset of disease compared to age and sex matched controls, because of an observed decreased incidence of cardiovascular diseases amongst these patients.15, 16. An explanation might be that physical activity is a risk factor for developing ALS, fuelled by anecdotal observations of famous athletes diagnosed with ALS, such as the 1930s American baseball player Lou Gehrig (to whom the disease was named after as an eponym in the United States).14 Other previously suggested associations between prior medical conditions and ALS risk are psychiatric disorders,17, 18 autoimmune diseases19, 20 and cancer21, 22. Which may indicate that there are subgroups of ALS with different pathophysiological mechanisms underlying the disease. Last, since dietary habits may have the potential to modify pathophysiological mechanisms in ALS, the role of diet as a risk factor for the development of ALS has been of particular interest. In the first part of this thesis we analyzed whether lifestyle factors, such as increased physical activity, co-occurring medical conditions and an altered dietary intake, are risk factors for sporadic ALS (chapters 2, 3 and 4).

Part II – EnvironmentEnvironmental risk factors can be ubiquitous and therefore important to a large proportion of the population. People can either be exposed to environmental risk factors through their occupations or in their residential area. For example, the first observations of an increased risk of ALS among agricultural workers date back more than 35 years and since then pesticides have frequently been suggested as the underlying occupational exposure responsible for the increased ALS risk.23, 24 Other environmental exposures of interest are for example metals,25-27 solvents,25, 28 air pollution,29, 30 and electric shocks or electromagnetic fields,31, 32 all of which have been inconsistently associated with ALS or neurodegenerative disorders in general. In these previous studies, an objective and quantitative method to assess environmental exposures is often lacking. In the second part of this thesis, we assessed the association of environmental factors, i.e. multiple occupational exposures, residential exposure to air pollution and residential exposure to electromagnetic fields, with the risk of developing ALS (chapters 5, 6 and 7). The exposure assessment was carried out by applying objective and valid tools, i.e. job exposure matrices (JEM) for occupational exposures33, 34 and land use regression (LUR) models for residential exposures.35, 36


13

Introduction

Part III - Genetics New causative genetic variants for ALS are discovered at an increasing pace, with the most frequently observed variants include the SOD1, FUS, TARDBP and C9orf72 genes.10 Despite the progress of gene discoveries in ALS, these genes explain only a minority of ALS cases. This can be partly explained by a large genetic heterogeneity in ALS, which is seen by locus heterogeneity, incomplete penetrance, multiple ALS mutations seen in one family (oligogenics) and genetic overlap with other diseases (pleiotropy).The genetic architecture of a complex trait or disease, such as ALS, is based on the allele frequency of disease variants, their effect sizes and the number of variants contributing to the risk. Few reports have shown that oligogenic inheritance within families does occur, with co-occurring variants in rare genes with large disease effect and more common susceptibility genes with intermediate disease effect.37 Furthermore, it has been argued that the distinction between familial and sporadic ALS is rather arbitrary,38 indicating that this oligogenic model may be applicable to all ALS cases. In addition, many ALS genes, but especially C9orf72, have marked genetic pleiotropy,meaning that it also causes other phenotypic traits in carriers of the mutation. C9orf72was initially discovered in ALS and frontotemporal dementia, but the gene has now been implicated in many different neurodegenerative and psychiatric diseases including Alzheimer’s disease, Parkinsonism, Huntington’s disease phenocopies, schizophrenia and bipolar disorder.39-42 Other ALS genes with a complex phenotyope are VCP, hnRNPA1 and hnRNPA2B1, which were discovered in multisystem proteinopathy (a disorder presenting as ALS, frontotemporal dementia, Paget’s disease of the bone or inclusion body myositis).43, 44

In the third part of this thesis, we analyzed the contribution of variants in hnRNPA1 and hnRNPA2B1 in Dutch patients with ALS, frontotemporal dementia and inclusion body myositis (chapter 8) and we analyzed the oligogenicity of apparently sporadic ALS including known ALS genes (chapter 9).


CHAPTER 1

14

REFERENCES

1. Huisman MHB, et al. Population based epidemiology of amyotrophic lateral sclerosis using capture-recapture methodology. Journal of Neurology Neurosurgery and Psychiatry. 2011; 82: 1165-70.

2. Pugliatti M, et al. Amyotrophic lateral sclerosis in Sardinia, insular Italy, 1995-2009. J Neurol. 2013; 260: 572-9.

3. Rooney J, et al. Survival analysis of irish amyotrophic lateral sclerosis patients diagnosed from 1995-2010. PLoS One. 2013; 8: e74733.

4. Phukan J, et al. The syndrome of cognitive impairment in amyotrophic lateral sclerosis: a population-based study. J Neurol Neurosurg Psychiatry. 2012; 83: 102-8.

5. Miller RG, et al. Riluzole for amyotrophic lateral sclerosis (ALS)/motor neuron disease (MND). Cochrane Database Syst Rev. 2012; 3: CD001447.

6. Logroscino G, et al. Incidence of amyotrophic lateral sclerosis in Europe. J Neurol Neurosurg Psychiatry. 2010; 81: 385-90.

7. Chio A, et al. Global epidemiology of amyotrophic lateral sclerosis: a systematic review of the published literature. Neuroepidemiology. 2013; 41: 118-30.

8. Ferraiuolo L, et al. Molecular pathways of motor neuron injury in amyotrophic lateral sclerosis. Nat Rev Neurol. 2011; 7: 616-30.

9. Byrne S, et al. Rate of familial amyotrophic lateral sclerosis: a systematic review and meta-analysis. J Neurol Neurosurg Psychiatry. 2011; 82: 623-7.

10. Renton AE, et al. State of play in amyotrophic lateral sclerosis genetics. Nat Neurosci. 2014; 17: 17-23.

11. Al-Chalabi A, et al. An estimate of amyotrophic lateral sclerosis heritability using twin data. Journal of Neurology Neurosurgery and Psychiatry. 2010; 81: 1324-6.

12. Brooks BR, et al. El Escorial revisited: revised criteria for the diagnosis of amyotrophic lateral sclerosis. Amyotroph Lateral Scler Other Motor Neuron Disord. 2000; 1: 293-9.

13. Armon C. Smoking may be considered an established risk factor for sporadic ALS. Neurology. 2009; 73: 1693-8.

14. Lewis M and Gordon PH. Lou Gehrig, rawhide, and 1938. Neurology. 2007; 68: 615-8.15. Sutedja NA, et al. Beneficial vascular risk profile is associated with amyotrophic lateral

sclerosis. J Neurol Neurosurg Psychiatry. 2011; 82: 638-42.16. TurnerMR, et al. Cardiovascular fitness as a risk factor for amyotrophic lateral sclerosis:

indirect evidence from record linkage study. J Neurol Neurosurg Psychiatry. 2012; 83: 395-8.17. Byrne S, et al. Aggregation of neurologic and neuropsychiatric disease in amyotrophic lateral

sclerosis kindreds: a population-based case-control cohort study of familial and sporadic amyotrophic lateral sclerosis. Ann Neurol. 2013; 74: 699-708.

18. Schreiber H, et al. Cognitive function in bulbar- and spinal-onset amyotrophic lateral sclerosis. A longitudinal study in 52 patients. J Neurol. 2005; 252: 772-81.

19. Turner MR, et al. Autoimmune disease preceding amyotrophic lateral sclerosis: an epidemiologic study. Neurology. 2013; 81: 1222-5.

20. Hemminki K, et al. Familial risks for amyotrophic lateral sclerosis and autoimmune diseases. Neurogenetics. 2009; 10: 111-6.

21. Freedman DM, et al. The association between cancer and amyotrophic lateral sclerosis. Cancer Causes Control. 2013; 24: 55-60.

22. Fois AF, et al. Cancer in patients with motor neuron disease, multiple sclerosis and Parkinson's disease: record linkage studies. J Neurol Neurosurg Psychiatry. 2010; 81: 215-21.

23. Rosati G, et al. Studies on epidemiological, clinical, and etiological aspects of ALS disease in Sardinia, Southern Italy. Acta Neurol Scand. 1977; 55: 231-44.

24. Gunnarsson LG, et al. An epidemic-like cluster of motor neuron disease in a Swedish county during the period 1973-1984. Neuroepidemiology. 1996; 15: 142-52.

25. Fang F, et al. Workplace exposures and the risk of amyotrophic lateral sclerosis. Environ Health Perspect. 2009; 117: 1387-92.


15

Introduction

26. Weisskopf MG, et al. Prospective study of occupation and amyotrophic lateral sclerosis mortality. Am J Epidemiol. 2005; 162: 1146-52.

27. Kamel F, et al. Lead exposure as a risk factor for amyotrophic lateral sclerosis. Neurodegener Dis. 2005; 2: 195-201.

28. Gunnarsson LG, et al. A case-control study of motor neurone disease: its relation to heritability, and occupational exposures, particularly to solvents. Br J Ind Med. 1992; 49: 791-8.

29. Ranft U, et al. Long-term exposure to traffic-related particulate matter impairs cognitivefunction in the elderly. Environ Res. 2009; 109: 1004-11.

30. Finkelstein MM and Jerrett M. A study of the relationships between Parkinson's disease and markersoftraffic-derivedandenvironmentalmanganeseairpollutionintwoCanadiancities.Environ Res. 2007; 104: 420-32.

31. HussA,etal.OccupationalexposuretomagneticfieldsandelectricshocksandriskofALS:The Swiss National Cohort. Amyotroph Lateral Scler Frontotemporal Degener. 2014: 1-6.

32. Frei P, et al. Residential distance to high-voltage power lines and risk of neurodegenerative diseases: a Danish population-based case-control study. Am J Epidemiol. 2013; 177: 970-8.

33. Sutedja NA, et al. Exposure to chemicals and metals and risk of amyotrophic lateral sclerosis: a systematic review. Amyotroph Lateral Scler. 2009; 10: 302-9.

34. Huss A, et al. Electric shocks at work in Europe: development of a job exposure matrix. Occup Environ Med. 2013; 70: 261-7.

35. Eeftens M, et al. Development of Land Use Regression models for PM(2.5), PM(2.5) absorbance, PM(10) and PM(coarse) in 20 European study areas; results of the ESCAPE project. Environ Sci Technol. 2012; 46: 11195-205.

36. Beelen R, et al. Development of NO2 and NOx land use regression models for estimating air pollution exposure in 36 study areas in Europe - The ESCAPE project. Atmospheric Environment. 2013; 72: 10-23.

37. van Blitterswijk M, et al. Evidence for an oligogenic basis of amyotrophic lateral sclerosis. Hum Mol Genet. 2012; 21: 3776-84.

38. Al-Chalabi A and Lewis CM. Modelling the effects of penetrance and family size on rates of sporadic and familial disease. Hum Hered. 2011; 71: 281-8.

39. DeJesus-Hernandez M, et al. Expanded GGGGCC hexanucleotide repeat in noncoding region of C9orf72 causes chromosome 9p-linked FTD and ALS. Neuron. 2011; 72: 245-56.

40. Beck J, et al. Large C9orf72 hexanucleotide repeat expansions are seen in multiple neurodegenerative syndromes and are more frequent than expected in the UK population. Am J Hum Genet. 2013; 92: 345-53.

41. Meisler MH, et al. C9orf72 expansion in a family with bipolar disorder. Bipolar Disord. 2013; 15: 326-32.

42. Lesage S, et al. C9orf72 repeat expansions are a rare genetic cause of parkinsonism. Brain. 2013; 136: 385-91.

43. Kim HJ, et al. Mutations in prion-like domains in hnRNPA2B1 and hnRNPA1 cause multisystem proteinopathy and ALS. Nature. 2013; 495: 467-73.

44. Watts GD, et al. Inclusion body myopathy associated with Paget disease of bone and frontotemporal dementia is caused by mutant valosin-containing protein. Nat Genet. 2004; 36: 377-81.



Meinie Seelen1*, Mark HB Huisman1*, Sonja W de Jong1, Kirsten RIS Dorresteijn1, Perry TC van Doormaal1, Anneke J van der Kooi2, Marianne de Visser2,

H Jurgen Schelhaas3, Leonard H van den Berg1# Jan H Veldink1#

1 Department of Neurology, Brain Center Rudolf Magnus, University Medical Center Utrecht, The Netherlands

2 Department of Neurology, Amsterdam Medical Center, University of Amsterdam, The Netherlands

3 Department of Neurology, Donders Institute for Brain, Cognition and Behaviour, Center for Neuroscience, Radboud University Nijmegen Medical Center,

The Netherlands

* These authors contributed equally to the manuscript# These authors contributed equally to the manuscript

Published in: Journal of Neurology, Neurosurgery and Psychiatry (2012)

Lifetime physical activity and the risk of

amyotrophic lateral sclerosis

CHAPTER 2

PART I - LIFESTYLE


CHAPTER 2

18

ABSTRACT

BackgroundIt has been hypothesized that physical activity is a risk factor for developing amyotrophic lateral sclerosis (ALS), fuelled by observations that professional soccer players and Gulf War veterans are at increased risk. In a population-based study, we determined the relation between physical activity and risk of sporadic ALS, using an objective approach for assessing physical activity.

Methods636 sporadic ALS patients and 2,166 controls, both population-based, completed a semi-structured questionnaire on lifetime history of occupations, sports and hobbies. To objectively compare energy cost of a lifetime history of occupational and leisure time physical activities and to reduce recall bias, metabolic equivalent scores were assigned to each activity based on the Compendium of Physical Activities.

ResultsALS patients had significantly higher levels of leisure time physical activity compared with controls (OR 1.08, 95% CI 1.02 to 1.14, p=0.008). No significant difference was found between patients and controls in the level of vigorous physical activities, including marathons and triathlons, or in occupational activity. Cumulative measures of physical activity in quartiles did not show a dose-response relationship.

ConclusionAn increased risk of ALS with higher levels of leisure time physical activity was found in the present study. The lack of association with occupational physical activity and the absence of a dose-response relationship strengthen the hypothesis that not increased physical activity per se, but rather a genetic profile or lifestyle promoting physical fitness increase ALS susceptibility.


19

Physical activity and ALS risk

INTRODUCTION

Sporadic amyotrophic lateral sclerosis (ALS) is believed to be a complex disease, with multiple genetic and environmental factors causing motor neuron degeneration.1 Ever since Lou Gehrig, the legendary 1930s baseball player known as 'The Iron Horse’, died from ALS, it has been hypothesized that physical activity is a risk factor for developing this disease. Although assuming an association based on an individual well-known patient is fraught with risk, the hypothesis has been fuelled by recent observations that professional soccer and football players, and Gulf War, veterans are at increased risk of sporadic ALS.2-

6 Several theories have been proposed that may explain the possible association of physical activity with ALS susceptibility.7-9

Although some studies have suggested a relation between physical activity and the risk of ALS, the results may have been biased due to methodological shortcomings, inherent in studying a relatively low-incidence disease.3, 10-13 A population-based case-control study can alleviate some of these limitations and, therefore, provide a high level of evidence in ALS exogenous risk factor studies.We performed a large population-based case control study in The Netherlands to determine the relation between physical activity and the risk of sporadic ALS, adjusted for known risk factors, using an objective quantitative approach for assessing physical activity, and taking into account the lifetime history of occupational and leisure time activities of each patient and control. To minimise recall bias, we measured the energy cost of the lifetime history of occupational and leisure time physical activities in an objective manner by assigning metabolic equivalent (MET) scores to each activity based on the Compendium of Physical Activities.14

METHODS

Study PopulationThe Prospective ALS study the Netherlands (PAN), is a population-based case-control study performed in the Netherlands during the period 1 January 2006 to 31 December 2010. Complete case ascertainment was ensured by continuous recruitment through multiple sources: neurologists, rehabilitation physicians, the Dutch Neuromuscular Patient Association and our ALS website.All patients diagnosed with possible, probable (laboratory supported) or definite ALS according to the revised El Escorial criteria were included.15 Medical records were scrutinised for eligibility of the patients, excluding patients with an ALS mimic syndrome or with a first, second or third degree family member with ALS. As exogenous factors - probably - had only a minor role in the development of ALS in patients with the highly penetrant C9orf72 repeat expansion, these patients, 43 in total, were excluded from our analysis.16-18


CHAPTER 2

20

To ascertain population-based controls, the general practitioner of the participating patient was asked to select individuals from his register in alphabetical order starting at the surname of the patient. The Dutch health care system ensures that every inhabitant is registered with a general practitioner, which makes this roster representative of the population. Controls were matched to the patients for gender and age (±5 years). This study, however, did not use individual matching, meaning that some general practitioners delivered several controls, while others delivered none. As can be seen in Table 2.1, our case and control groups were well matched for age and gender. Blood relatives or spouses of patients were not eligible to be controls, to prevent overmatching. Ethics approval was provided by the institutional review board of the University Medical Centre Utrecht. All participants gave written informed consent.

Data collectionA structured questionnaire was used to collect demographic and clinical characteristics of participants and to obtain data regarding lifetime physical activities. Participants were asked to recollect all their jobs and to describe the various activities they had to perform during these jobs. They were also asked to list all their leisure time activities, consisting of sports and hobbies. For each activity, the participant was asked to state the number of years and how many hours per week the activity was performed. Specific questions were asked about vigorous physical activities (eg, marathon, triathlon, etc.). This questionnaire was part of a larger questionnaire containing questions on several other exogenous factors. Participants were, therefore, blinded to the hypothesis being tested. In the patient group, only data referring to the period before disease onset were analysed. Survival status of patients was recorded up to 8 August 2011 and obtained through the municipal personal records database or from the general practitioner. If the questionnaire was not completed in full or if data were found to be inconsistent, participants were approached by telephone to complete or correct the data. To ensure blinding, all questionnaires were coded prior to processing and analysis.

Classification of physical activitiesTo objectively quantify the cumulative lifetime physical activity level of participants, all reported activities were scored and coded based on the Compendium of Physical Activities.14 The Compendium provides a coding scheme that links specific activities performed in various settings with their respective MET. The definition of a MET is the ratio of work metabolic rate to a standard resting metabolic rate. A MET score of 1.0 (i.e. the standard or resting metabolic rate while sitting quietly) is defined as 1 kcal × kg-1 body weight × h-1. MET levels for specific activities, as reported in the Compendium, were established by reviewing published and unpublished studies that measured the energy cost of human physical activities. The compendium describes 605 specific activities. Assignment of MET scores to the activities enabled us to calculate cumulative scores of all reported physical activities:


21


where k represents an activity from the lifetime job or leisure time history. Because of the magnitude of the cumulative score, it was divided by 1000. Activities that had a MET score of ≤ 1.5 (eg, listening to music, reading, playing chess, needlework) were not included in the analysis. Subsequently, military service (not occupation) or periods spent as a homemaker were excluded because of difficulties quantifying these activities. Military service was mandatory for male study participants, during a 15 to 24 month period around the age of 18 years and will therefore have minimal influence on total cumulative physical activity. In our study, 34% of patients compared with 35% of controls joined the military service (χ2 test: p=0.73), and 12% of both patients and controls reported periods spent as a homemaker (χ2 test: p=0.77).

Statistical methodsUnivariate and multivariate logistic regression were used to determine the association of physical activity and the risk of ALS. Standard unconditional logistic regression was used as the study did not include individual case-control pairs but was frequency-matched. The risk of ALS with cumulative scores of physical activity was analysed separately for leisure time activity, occupational activity and total activity (the combined leisure time and occupational activity) as a continuous variable. Furthermore, to determine a dose-response relationship, physical activity was categorised into quartiles based on the data of controls. The first quartile with the lowest intensity in physical activities was defined as the reference category. Multivariate logistic regression was used to determine the association between the four levels of physical activity and ALS. A separate multivariate logistic regression analysis was performed to determine the effect of vigorous physical activity (ever/never) on the risk of ALS. OR and p values were derived from these analyses. In the multivariate model, the ORs were adjusted for gender, age (at onset for patients and at the date the questionnaire was completed for controls), level of education (divided into seven categories ranging from no education to university), premorbid body mass index, current alcohol consumption and current smoking. In patients, current alcohol consumption and current smoking were determined at the time of disease onset, so before diagnosis and before the questionnaire was completed.To determine a difference in the maximum intensity of the activities performed, the maximum MET scores were calculated (excluding the duration in years or the hours per week) and analysed using the Mann Whitney U test.

A Cox regression analysis was performed to determine whether survival of patients was associated with physical activity. Survival was defined as the time from symptom onset to death or to the censoring date of 8 August 2011. The HR derived from these analyses were adjusted for gender, age at onset, site of onset and current smoking. Physical activity was entered into the model as a continuous variable. The same method was used to


CHAPTER 2

22

determine the effect of physical activity on the age at onset of ALS patients, adjusting for gender and site of onset. To adjust appropriately for age, an interaction term of diagnosis and physical activity was introduced into the Cox regression analysis using age at time of completing the questionnaire for controls. In the above mentioned models, we performed a complete case analysis, using only those cases without any missing values. A Bonferroni correction for multiple testing was applied adjusting for three tests (leisure time, occupational and total activity), a p value of 0.05/3=0.017 was considered significant.

RESULTS

In the population-based study, 636 (84%) of the 760 patients who gave informed consent to participate in the study between 1 January 2006 and 31 December 2010, returned the questionnaire. Of the 2332 population-based controls who gave informed consent, 93% returned their questionnaires (2166 controls). Table 2.1 shows the characteristics of 636 patients and 2166 controls. The patient characteristics of the responders and the non-responders were similar. Of the 2802 participants, 2281 (81.4%) had completed the questionnaires on physical activities without any missing values in duration in years or hours per week. The distributions for gender, age at onset and site of onset in ALS patients were similar to those previously reported in population-based studies.19

A greater amount of leisure time physical activity was associated with an increased risk of ALS in the present study (adjusted OR 1.08, p=0.008) (Table 2.2).

ControlsALS patients

Mea

n ac

tivity

sco

re

2.0

1.5

1.0

0.5

0.0

Mean difference: p = 0.004

Figure 2.1 Mean leisure time activity of ALS patients and controls. Patients mean = 1.51, 95% CI 1.30 to 1.72, controls mean = 1.25, 95% CI 1.18 to 1.32. ALS, amyotrophic lateral sclerosis.


23


This is also illustrated in Figure 2.1, showing the mean cumulative scores for leisure time activity (patient mean=1.51, 95% CI 1.30 to 1.72; control mean=1.25, 95% CI 1.18 to 1.32; p=0.004). Occupational and total physical activity were not associated with the risk of ALS (Table 2.2); no dose-response relationship was seen with physical activity (Figure 2.2) and none of the vigorous physical activities showed a significant association with ALS (Table 2.3). Maximum MET scores did not differ significantly between ALS patients and controls, implying that there was no difference in the maximum intensity of activities (all p values >0.35, not shown).Survival analyses showed that none of the cumulative measures of physical activity was associated with survival (all p values >0.10). Of 636 patients, 63% died before the censoring date of 8 August 2011. The cumulative measures of leisure time, occupational and total activity did, however, show a significant relation with age at onset (all HR 0.94 to 0.95,

Table 2.1 Baseline demographic and clinical characteristics of participants

ALS patients Controls

Characteristic (n = 636) (n = 2166) p Value

Male (n (%)) 395 (62.1) 1259 (58.1) 0.17

Age (years) (median (range))* 63 (23 to 87) 62 (20 to 91) 0.91

Site of onset (n (%))

Bulbar 204 (32.3)

Spinal 427 (67.7)

El Escorial classification (n (%))

Definite 112 (17.8)

Probable 280 (44.6)

Probable lab supported 111 (17.7)

Possible 119 (18.9)

Education (n (%))

No education 2 (0.3) 3 (0.1)

Primary school 54 (8.5) 131 (6.1)

Junior vocational education 127 (20.0) 356 (16.5)

Lower general secondary education 149 (23.4) 474 (21.9) 0.02

Intermediate vocational education 106 (16.7) 410 (18.9)

Higher general secondary education 45 (7.1) 186 (8.6)

College/University 153 (24.1) 604 (27.9)

BMI (kg/m2) (median (range)) 24.1 (12 to 48) 25.6 (16 to 53) <0.001

Current smoking (n (%)) 133 (20.9) 288 (13.3) <0.001

Current alcohol consumption (n (%)) 475 (74.7) 1846 (85.3) <0.001

ALS, amyotrophic lateral sclerosis; BMI, body mass index.* Age at onset in patients, and age on which questionnaire was completed in controls.


CHAPTER 2

24

p values ≤0.009). In order to show whether this effect was specific for patients or valid for age at the time of the questionnaire for controls, two additional analyses were performed: (1) an interaction term of diagnosis and physical activity was introduced into the model (all p values >0.45), and (2) a multivariate Cox regression was performed in controls using questionnaire completion as the event (p≤0.002). Both indicated that the relationship between physical activity and age at onset was an age related effect and thus not disease related. Kaplan-Meier curves of total activity of both survival and age at onset are shown in the online Supplementary Figure S2.1.

Table 2.2 Odds ratios for the relationship between ALS and the cumulative scores of physical activity

Crude OR Adjusted OR*

Variable (95% CI) p Value† (95% CI) p Value†

Leisure time activity 1.08 (1.02 to 1.13) 0.005 1.08 (1.02 to 1.14) 0.008

Occupational activity 1.02 (0.99 to 1.06) 0.19 1.00 (0.96 to 1.04) 0.90

Total activity 1.03 (0.99 to 1.06) 0.12 1.02 (0.98 to 1.06) 0.30

ALS, amyotrophic lateral sclerosis.* Adjusted for gender, age, body mass index, current smoking, current alcohol consumption and level of education. † Bonferroni adjusted p values of < 0.017 (0.05/3) were considered significant.

1.00

0.72

1.01 1.12

1.00

0.81

0.95 0.99 1.00

0.95

0.82

0.97

0.4

0.6

0.8

1.0

1.2

1.4

1.6

OR

Leisure time activity Occupational activity Total activity

Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4

Figure 2.2 Odds ratios (OR) with 95% confidence intervals for the relationship between quartiles of leisure time, occupational and total activity and the risk of ALS. ORs were adjusted for gender, age at onset, body mass index, current smoking, current alcohol consumption and level of education. The physical activity score was categorized into quartiles (Q) based on the data of controls. Q1, 1st quartile; Q2, 2nd quartile; Q3, 3rd quartile; Q4, 4th quartile.


25


DISCUSSION

Evidence for an increased risk of ALS with higher levels of leisure time physical activity is provided by the present population-based case-control study. Occupational physical activity and performing vigorous physical activities, however, do not appear to modify ALS susceptibility in this study. The discrepancy between leisure time and occupational physical activity strengthens the hypothesis that physical activity itself is not causative per se but that being athletic is a phenotypic expression of a genetic profile, mediated by exogenous factors, that increases the risk of ALS.20-23 Our observation that none of the physical activity measures was related to age at onset or survival further supports this hypothesis.Two systematic reviews on the association between ALS and physical activity concluded that there is a consistent pattern of well-designed studies showing no link between physical activity and sporadic ALS.11, 12 The best evidence available at that time was provided by a single population based case-control study that showed no association.24 After publication of these reviews, however, a small but well-designed European population-based pilot case-control study identified an increased risk of ALS with higher levels of physical activity.13 In concordance with these conflicting results, a third and the most recent, systematic review concluded that current evidence for physical activity as a risk factor in motor neuron disease is not of sufficient caliber to allow undisputed conclusions.8 The conflicting results found in studies on the association between physical activity and ALS, may partly be due to differences in methodological design. These differences concern: (1) the blinding of interviewers to disease status of respondents or the hypotheses being tested; (2) referral bias, which was common with cases often ascertained at specialist clinics; (3) adjustment for confounders, which was not carried out in all analyses; and (4) the method of assessing physical activity, which in most studies was susceptible to recall bias.8, 11, 12 Recall bias is due to differential recall of past exposures between patients and controls. As ALS patients actively search for an explanation of their disease or may have an assumption about the underlying cause, case-control studies in ALS using questionnaires

Table 2.3 Vigorous physical activities among ALS patients and controls

ALS patients Controls Adjusted OR*

Variable (n = 635) (n = 2167) (95% CI) p Value†

Vigorous physical activity (n (%)) 103 (16) 296 (14) 1.24 (0.96 to 1.61) 0.10

Marathon 12 (1.9) 32 (1.5) 1.15 (0.58 to 2.29) 0.69

Triathlon 3 (0.5) 6 (0.3) 1.21 (0.29 to 4.98) 0.80

Ice skating tours >200km 7 (1.1) 18 (0.8) 1.35 (0.54 to 3.37) 0.52

ALS, amyotrophic lateral sclerosis. * Adjusted for gender, age, body mass index, current smoking, current alcohol consumption and level of education. † Bonferroni adjusted p values of < 0.017 (0.05/3) were considered significant.


CHAPTER 2

26

are prone to this bias. Our study was designed to minimize the risk of recall and referral bias. First, recall bias was reduced by using the Compendium of Physical Activities14 to quantify objectively physical activity based on type of occupation or type of leisure time activities, instead of directly asking participants how physically active they have been in their life or during the listed activities. As the questionnaire on leisure time and occupational activities was part of a more comprehensive questionnaire, participants were blinded to the study hypothesis, which further reduced the risk of recall bias. Interviewers, who called participants to complete returned questionnaires, were also unaware of the hypothesis being tested. Referral bias may occur when patients are ascertained from tertiary care centres. It has been demonstrated that ALS patients attending these referral centres do not represent a random sample of all ALS patients.25, 26 A difference in physical activity levels of these patients compared with non-referred patients will lead to biased results. The population-based design using multiple sources to ensure complete case ascertainment minimized the risk of referral bias in the present study, which was strengthened by the observation that the demographics of the patients in our study resembled those of patients in other population-based studies.19, 27, 28

We acknowledge certain limitations of the present study: 18.6% of the participants had at least one missing value for the duration of, or the hours per week spent on, one of the listed activities, even after being called by an interviewer to complete the returned questionnaire. This is probably the result of the level of detail of the questionnaire concerning past events. The fact that this information was so elaborate, however, enabled us to precisely quantify lifetime energy expenditure during leisure time and occupational activities. Also, it is noteworthy that ALS patients had significantly less higher education (p<0.02), which is congruent with a previous observation that there is a preponderance among ALS patients of blue-collar jobs for which a higher level academic education is often not required. Nevertheless, our controls may have been better educated as people with a higher education tend to participate in scientific surveys more readily.29 The effects, however, of this observation will have been minimal as we adjusted all analyses for education. Further, we acknowledge that the quantification of the lifetime energy expenditure is still an estimate of real energy expenditure. A study, however, in which these data are prospectively being collected will probably not be feasible in a low incidence disease such as ALS. Finally, although our study was designed to maximize blinding of the participants to the hypothesis of the study, we cannot exclude the fact that a proportion of the patients may have been aware of the theory of physical activity as a possible risk factor, which may have been a source of residual recall bias.Our finding that an increased leisure time physical activity is related to an increased risk of ALS but occupation activity is not, raises doubts regarding the role of physical activity in causing ALS. Because of existing cellular and genetic evidence supporting the biological plausibility of the association, some have suggested that physical activity is indeed causative.8, 30, 31 Several genes associated with the response to exercise (i.e. ciliary neurotrophic factor, leukaemia inhibitory factor and vascular endothelial growth factor


27


2) have been identified as possible modifiers of ALS susceptibility.32-34 Also, oxidative stress and glutamate excitotoxicity are considered candidate mechanisms to link ALS and physical activity.7, 8, 35 The biologically plausible link between physical activity and ALS has been carefully reviewed.8

Biological plausibility alone, however, does not prove causation. Useful time-tested criteria for determining whether an association is causal are designed by Bradford Hill.36, 37 The Bradford-Hill criteria include strength, consistency, specificity, temporality, dose-response relation, plausibility, coherence, experiment and analogy. The associations found in the present study do not meet most of these criteria. First, strength. If an association is weak, it is more plausible that underlying actual causative factors that go hand-in-hand with the studied factor are in fact responsible for the observed association. In our study, if physical activity were causative, an increase in physical activity of 10 000 MET, which can be provided by 50 years of 50 hours cycling per week for example, would be associated with an increase of odds of developing ALS of only 2.2. Further, if we had applied a more stringent threshold that also corrects for the analyses on vigorous physical activities (threshold p=0.05/7=0.007), the association (p=0.008) would not even have been significant, further emphasising the weakness of the association. Second, consistency. A real causative factor is more likely to be repeatedly observed in different studies, using different methodologies and performed in different places, circumstances and times. Previous studies, as already emphasized, have shown large inconsistencies, and even within the present study there is an inconsistency between occupational and leisure time physical activity.11-13, 24 Finally, the absence of a dose-response relation also does not support the notion that causation is the most likely interpretation of the association between leisure time physical activity and ALS. Recent findings of a beneficial vascular risk profile in both patients and their relatives,6 a reduced frequency of coronary heart disease premorbidly in ALS,22, 38 and an increased risk of ALS with physical fitness, but not muscle strength,21 further indicate that a common factor underlies both physical/cardiovascular fitness and risk of ALS.39 A genetic profile, therefore, modified by exogenous factors, that both promotes physical fitness and increases ALS susceptibility might be a more credible explanation for the associations between physical activity and ALS.20, 22

In conclusion, the present population-based case-control study strengthens this hypothesis. Identifying genetic, developmental and environmental factors that contribute to physical fitness may provide a worthwhile lead in unravelling the pathophysiological mechanisms in ALS.


CHAPTER 2

28

REFERENCES

1. Kiernan MC, et al. Amyotrophic lateral sclerosis. Lancet. 2011; 377: 942-55.2. Lehman EJ, et al. Neurodegenerative causes of death among retired National Football League

players. Neurology. 2012; 79: 1970-4.3. Chio A, et al. Severely increased risk of amyotrophic lateral sclerosis among Italian professional

football players. Brain. 2005; 128: 472-6.4. Weisskopf MG, et al. Prospective study of military service and mortality from ALS. Neurology.

2005; 64: 32-7.5. Chio A, et al. ALS in Italian professional soccer players: the risk is still present and could be

soccer-specific.Amyotroph Lateral Scler. 2009; 10: 205-9.6. Huisman MH, et al. Family history of neurodegenerative and vascular diseases in ALS: a

population-based study. Neurology. 2011; 77: 1363-9.7. Longstreth WT, et al. Hypotheses to explain the association between vigorous physical activity

and amyotrophic lateral sclerosis. Med Hypotheses. 1991; 34: 144-8.8. Harwood CA, et al. Physical activity as an exogenous risk factor in motor neuron disease

(MND): a review of the evidence. Amyotroph Lateral Scler. 2009; 10: 191-204.9. Vanacore N, et al. Job strain, hypoxia and risk of amyotrophic lateral sclerosis: Results from a

deathcertificatestudy.Amyotroph Lateral Scler. 2010; 11: 430-4.10. Armon C. An evidence-based medicine approach to the evaluation of the role of exogenous

risk factors in sporadic amyotrophic lateral sclerosis. Neuroepidemiology. 2003; 22: 217-28.11. Veldink JH, et al. Physical activity and the association with sporadic ALS. Neurology. 2005;

64: 241-5.12. Armon C. Sports and trauma in amyotrophic lateral sclerosis revisited. J Neurol Sci. 2007;

262: 45-53.13. Beghi E, et al. Amyotrophic lateral sclerosis, physical exercise, trauma and sports: results of a

population-based pilot case-control study. Amyotroph Lateral Scler. 2010; 11: 289-92.14. Ainsworth BE, et al. Compendium of physical activities: an update of activity codes and MET

intensities. Med Sci Sports Exerc. 2000; 32: S498-504.15. Brooks BR, et al. El Escorial revisited: revised criteria for the diagnosis of amyotrophic lateral

sclerosis. Amyotroph Lateral Scler Other Motor Neuron Disord. 2000; 1: 293-9.16. Renton AE, et al. A hexanucleotide repeat expansion in C9orf72 is the cause of chromosome

9p21-linked ALS-FTD. Neuron. 2011; 72: 257-68.17. DeJesus-Hernandez M, et al. Expanded GGGGCC hexanucleotide repeat in noncoding region

of C9orf72 causes chromosome 9p-linked FTD and ALS. Neuron. 2011; 72: 245-56.18. Ishiura H, et al. C9orf72 repeat expansion in amyotrophic lateral sclerosis in the Kii peninsula

of Japan. Arch Neurol. 2012; 69: 1154-8.19. Logroscino G, et al. Incidence of amyotrophic lateral sclerosis in Europe. J Neurol Neurosurg

Psychiatry. 2010; 81: 385-90.20. Scarmeas N, et al. Premorbid weight, body mass, and varsity athletics in ALS. Neurology. 2002;

59: 773-5.21. Mattsson P, et al. Physical fitness, but not muscle strength, is a risk factor for death in

amyotrophic lateral sclerosis at an early age. J Neurol Neurosurg Psychiatry. 2012; 83: 390-4.22. TurnerMR, et al. Cardiovascular fitness as a risk factor for amyotrophic lateral sclerosis:

indirect evidence from record linkage study. J Neurol Neurosurg Psychiatry. 2012; 83: 395-8.23. ChioAandMoraG.Physicalfitnessandamyotrophiclateralsclerosis:dangerousliaisonsor

common genetic pathways? J Neurol Neurosurg Psychiatry. 2012; 83: 389.24. Longstreth WT, et al. Risk of amyotrophic lateral sclerosis and history of physical activity: a

population-based case-control study. Arch Neurol. 1998; 55: 201-6.25. Lee JR, et al. Prognosis of amyotrophic lateral sclerosis and the effect of referral selection. J

Neurol Sci. 1995; 132: 207-15.26. Sorenson EJ, et al. Effect of referral bias on assessing survival in ALS. Neurology. 2007; 68: 600-2.


29


27. McGuire V, et al. Incidence of amyotrophic lateral sclerosis in three counties in western Washington state. Neurology. 1996; 47: 571-3.

28. Forbes RB, et al. The incidence of motor nueron disease in Scotland. J Neurol. 2007; 254: 866-9.29. Galea S and Tracy M. Participation rates in epidemiologic studies. Ann Epidemiol. 2007; 17:

643-53.30. Turner MR, et al. Concordance between site of onset and limb dominance in amyotrophic

lateral sclerosis. J Neurol Neurosurg Psychiatry. 2011; 82: 853-4.31. Ferraiuolo L, et al. Transcriptional response of the neuromuscular system to exercise training

and potential implications for ALS. J Neurochem. 2009; 109: 1714-24.32. LambrechtsD,etal.VEGFisamodifierofamyotrophiclateralsclerosisinmiceandhumans

and protects motoneurons against ischemic death. Nat Genet. 2003; 34: 383-94.33. Zheng C, et al. Vascular endothelial growth factor prolongs survival in a transgenic mouse

model of ALS. Ann Neurol. 2004; 56: 564-7.34. Al-ChalabiA,etal.Ciliaryneurotrophicfactorgenotypedoesnotinfluenceclinicalphenotype

in amyotrophic lateral sclerosis. Ann Neurol. 2003; 54: 130-4.35. Ilieva EV, et al. Oxidative and endoplasmic reticulum stress interplay in sporadic amyotrophic

lateral sclerosis. Brain. 2007; 130: 3111-23.36. Hill AB. The Environment and Disease: Association or Causation? Proc R Soc Med. 1965; 58:

295-300.37. Gallo V, et al. Smoking and risk for amyotrophic lateral sclerosis: analysis of the EPIC cohort.

Ann Neurol. 2009; 65: 378-85.38. Sutedja NA, et al. Beneficial vascular risk profile is associated with amyotrophic lateral

sclerosis. J Neurol Neurosurg Psychiatry. 2011; 82: 638-42.39. Wicks P. Hypothesis: higher prenatal testosterone predisposes ALS patients to improved

athletic performance and manual professions. Amyotroph Lateral Scler. 2012; 13: 251-3.

SUPPLEMENTAL MATERIAL

Age at onset (years)80604020

Affe

cted

(%)

100

80

60

40

20

0

high levellow level

A)

Survival (years)151050

Aliv

e (%

)

100

80

60

40

20

0

high levellow level

B)

Age at onset (years)80604020

Affe

cted

(%)

100

80

60

40

20

0

high levellow level

A)

Survival (years)151050

Aliv

e (%

)

100

80

60

40

20

0

high levellow level

B)

Figure S2.1 Kaplan-Meier curves comparing high (dashed line) versus low (solid line) level total activity in relation to (A) age at onset and (B) survival. Log Rank test for age at onset, p = 0.51, and survival, p = 0.77.



Prior medical conditions and

the risk of amyotrophic lateral sclerosis

CHAPTER 3

Journal of Neurology (2014)

Meinie Seelen1, Perry TC van Doormaal1, Anne E Visser1, Mark HB Huisman1, Margot HJ Roozekrans1, Sonja W de Jong1, Anneke J van der Kooi2, Marianne de Visser2,

Nicol C Voermans3, Jan H Veldink*1, Leonard H van den Berg*1

1 Department of Neurology, Brain Center Rudolf Magnus University Medical Center Utrecht, The Netherlands

2 Department of Neurology, Academic Medical Center University of Amsterdam, The Netherlands

3 Department of Neurology, Donders Institute for Brain, Cognition and BehaviorCenter for Neuroscience, Radboud University Nijmegen Medical Center, The Netherlands

* These authors contributed equally to this work


CHAPTER 3

32

ABSTRACT

Sporadic amyotrophic lateral sclerosis (ALS) is believed to be a complex disease in which multiple exogenous and genetic factors interact to cause motor neuron degeneration. Elucidating the association between medical conditions prior to the first symptoms of ALS could lend support to the theory that specific subpopulations are at risk of developing ALS and provide new insight into shared pathogenic mechanisms. We performed a population-based case-controls study in The Netherlands, including 722 sporadic ALS patients and 2268 age and gender matched controls. Data on medical conditions and use of medication were obtained through a structured questionnaire. Multivariate analyses showed that hypercholesterolemia (OR 0.76, 95% CI 0.63-0.92, P = 0.006), the use of statins (OR 0.45, 95% CI 0.35-0.59, P = 1.86 x 10-9) or immunosuppressive drugs (OR 0.26, 95% CI 0.08-0.86, P = 0.03) were associated with a decreased risk of ALS. Head trauma was associated with an increased ALS susceptibility (OR 1.95, 95% CI 1.11-3.43, P = 0.02). No association was found with autoimmune diseases, cancer, psychiatric disorders or cardiovascular diseases, or survival. The lower frequency of hypercholesterolemia and less use of statins in ALS patients indicate a favorable lipid profile prior to symptom onset in at least a subpopulation of ALS. Prior head trauma is a risk factor for ALS and the significantly lower use of immunosuppressive drugs in ALS patients could suggest a protective effect. The identification of specific subpopulations at risk for ALS may provide clues towards possible pathogenic mechanisms.

INTRODUCTION

Amyotrophic lateral sclerosis (ALS) is a devastating neurodegenerative disease in which the vast majority of cases appear to be sporadic (approximately 90-95%). Sporadic ALS is believed to be a complex disease in which multiple exogenous and genetic factors interact to cause motor neuron degeneration.1 Elucidating the association between medical conditions and medication use prior to the first symptoms of ALS could lend support to the theory that specific subpopulations are at risk of developing ALS and provide new insight into shared pathogenic mechanisms.Associations with increased or decreased risk of developing ALS have been reported for cardiovascular diseases,2-4 psychiatric disorders,5, 6 cancer,7, 8 autoimmune diseases9, 10 and (head) trauma11, 12 preceding ALS onset, though sometimes with conflicting results. To obtain a high level of evidence for the association of ALS and a wide range of prior medical conditions and use of medication, we performed a large population-based, case-control study in the Netherlands including incident ALS cases and gender- and age-matched controls.


33

Prior medical conditions and ALS risk

METHODS

Study PopulationBetween January 1st, 2006 and March 31st, 2011, we performed a population-based case-control study entitled “Prospective ALS study the Netherlands” (PAN). The Netherlands provides ideal circumstances for a population-based study. It is a densely populated country with healthcare which is accessible to all inhabitants and a well-developed infrastructure.13 Complete acquisition of all patients with ALS was ensured by continuous recruitment through multiple sources: neurologists, rehabilitation physicians, the Dutch Neuromuscular Patient Association and our ALS website. All patients newly diagnosed as possible, probable (laboratory supported) or definite ALS according to the revised El Escorial Criteria were included.14 Excluded were patients who had a first, second or third degree family member with ALS, defined as familial ALS. In order to obtain population-based controls, the general practitioner of the participating patient was asked to select individuals from his register in alphabetical order starting at the surname of the patient. In The Netherlands, the healthcare system ensures that every inhabitant is registered at a general practitioner, which makes this list representative of the population. The controls were frequency-matched to the patients for gender and age (plus or minus five years). Spouses or blood-relatives of the patients were not eligible to be controls to prevent overmatching. We determined the C9orf72 repeat in all ALS patients and in one-third of controls, by performing a repeat-primed PCR reaction as described previously.15, 16

Data Collection A structured questionnaire was used to collect clinical characteristics of the participants and to obtain data regarding medical history and premorbid use of medication. In these questionnaires, participants were asked to recollect their entire medical history (including the year of the medical event), and their past use of medication (including the years they used the medication). In the patient group, only the data referring to the period before symptom onset were analyzed. All questionnaires were checked thoroughly for missing data or inconsistencies, and participants were approached by telephone to complete or correct the data. To ensure blinding, all questionnaires were coded prior to processing and analysis. The survival status of patients was obtained by checking the municipal population register and/or contacting the general practitioner on a 3-monthly basis for both patients and controls.

Classification of DataMedical history was categorized into eight main groups based on expert opinion: cardiovascular diseases, neurodegenerative diseases, psychiatric disorders, cancer, infectious diseases, autoimmune diseases, trauma and surgery. The use of medication was scored and coded using the Anatomical Therapeutic Chemical (ATC) classification system of the World Health Organization Collaborating Centre for Drug Statistics Methodology.17


CHAPTER 3

34

One of the major purposes of the ATC coding system is to enable the presentation and comparison of medication consumption statistics internationally. The ATC classification system divides medication into 14 main groups according to the organ or system on which it acts. Further classification into 4 levels of subgroups is based on the pharmacological, therapeutic and chemical properties of the medication. The assignment of the ATC codes enabled us to compare premorbid use of medication at different levels based on an internationally accepted classification system.

Statistical AnalysisDifferences in baseline characteristics were evaluated using the χ2 test for categorical variables and the Mann-Whitney U test for continuous variables. A multivariate logistic regression model was used to determine the association between medical history, premorbid use of medication and the risk of ALS. The odds ratios (OR) and 95% confidence intervals (CI) derived from these analyses were adjusted for gender, age (at onset for patients and at the date the questionnaire was completed for controls), education (three levels: elementary school, middle/high school and college/university), current smoking and current alcohol consumption (current meaning at onset for patients and at the time the questionnaire was completed for controls). Since C9orf72 repeat expansions were discovered to be an important factor in ALS causation and patients with a C9orf72 repeat expansion may represent a subgroup with different lifestyle and environmental factors, we performed a separate logistic regression analysis excluding apparently sporadic ALS patients with a C9orf72 repeat expansion. Subsequently, we also performed a logistic regression analysis in the subgroup of patients with a C9orf72 repeat expansion.To determine whether survival of patients was associated with medical history and premorbid use of medication, a Cox proportional hazard model was used. The hazard ratios (HR) derived from these analyses were adjusted for gender, age at onset and site of onset. The same method was used to evaluate the effect of medical history and premorbid use of medication on the age at onset of ALS patients, adjusted for gender and site of onset.

RESULTS

Clinical Characteristics In this population-based study, 722 (83%) of 867 sporadic ALS patients and 2268 (95%) of 2454 controls, returned the questionnaire. Table 3.1 presents the characteristics of patients and controls. The patient characteristics of the responders and the non-responders were similar. The distribution of age at onset, gender and site of onset in ALS patients were found to be similar to those previously reported in Europe.18


35


Medical Conditions Prior to ALS Onset 575 patients (80%) compared to 1850 controls (82%) reported one or multiple medical conditions prior to ALS onset (P = 0.25). Table 3.2 shows the association of medical conditions with the risk of developing ALS in multivariate analyses. Hypercholesterolemia was associated with a decreased risk (OR 0.76, 95% CI 0.63-0.92, P = 0.006, Figure 3.1). Because all other cardiovascular diseases, except stroke, showed ORs < 1.0, we also analyzed the association between ever having had any of these cardiovascular diseases and the risk of ALS which showed no association (OR 0.85, 95% CI 0.72-1.02, P = 0.08). When we took current smoking and current alcohol consumption out of the multivariate model, the ORs remained similar (data not shown). However, when we added body mass index (BMI) to the model, the association with hypercholesterolemia was no longer significant (OR 0.84, 95% CI 0.69-1.02, P = 0.08), indicating that BMI is either a confounder or a factor in the same cascade with hypercholesterolemia.Trauma was associated with an increased risk of developing ALS (OR 1.36, 95% CI 1.03-1.79, P = 0.03), which was mainly caused by the contribution of head trauma (i.e. skull

Table 3.1 Demographic and clinical characteristics of participants

ALS patients(n = 722)

Controls(n = 2268)

P Value

Characteristic

Male, n (%) 435 (60.2) 1316 (58.0) 0.29

Age, y, median (IQR)a 62.8 (56.7-69.6) 63.1 (57.3-69.9) 0.29

Bulbar site of onset, n (%) 237 (32.8)

El Escorial classification, n (%)

Definite 117 (17.2)

Probable 295 (43.4)


Possible 128 (18.9)

Education, n (%)

No education 3 (0.4) 5 (0.2)

Elementary school 65 (9.0) 142 (6.3) 0.02

Middle school/High school 478 (66.2) 1481 (65.4)

College/University 176 (24.4) 637 (28.1)

Current smoking, n (%) 144 (19.9) 294 (13.0) <0.001

Current alcohol consumption, n (%) 527 (73.0) 1925 (84.9) <0.001

BMI, kg/m2, median (IQR) 24.1 (22.1-26.5) 25.6 (23.6-27.8) <0.001

C9orf72 repeat expansion, n (%)b 43 (5.9) 0 (0.0) <0.001

Abbreviations: ALS = amyotrophic lateral sclerosis; BMI = body mass index. a Age at onset in patients, and age at which questionnaire was completed in controls. b C9orf72 repeats were analyzed in all 722 ALS patients and in 762 controls (33.6%).


CHAPTER 3

36

fractures, concussion or intracranial hemorrhage; OR 1.95, 95% CI 1.11-3.43, P = 0.02). Taking current smoking and current alcohol consumption out of the multivariate analyses, or adding BMI, did not essentially alter the ORs (data not shown). None of the other categories was associated with ALS susceptibility. Other diseases that have previously been linked to ALS risk,5-10 such as Parkinson’s disease (Table 3.2), psychotic illness (Table 3.2), various autoimmune diseases and several cancer types (Table 3.3), were analyzed but no associations were found.

Medication Use Prior to ALS Onset 393 patients (54%) and 1297 controls (57%) used one or multiple medications at onset of disease for patients or at time of completing the questionnaire for controls (P = 0.19). Table 3.4 shows the ATC-coded main groups. Medication of the ‘cardiovascular system’ was associated with a decreased ALS susceptibility (OR 0.79, 95% CI 0.65-0.96, P = 0.02). Analyzing the ATC-coded subgroups showed that this association was due to the use of lipid-modifying agents (statins) (OR 0.45, 95% CI 0.35-0.59, P = 1.86 x 10-9, Table S3.1, Figure 3.1). Removing current smoking and current alcohol consumption from the multivariate analyses did not change the association with ALS risk; adding BMI led to a non-significant association with medication of the ‘cardiovascular system’ (OR 0.95, 95% CI 0.78-1.16, P = 0.61), and to a modest increase in the OR for statin use (OR 0.51, 95% CI 0.39-0.61, P = 4.84 x 10-7).The main group ‘antineoplastic and immunomodulating agents’ was associated with a decreased risk of ALS (OR 0.35, 95% CI 0.14-0.91, P = 0.03). Analyzing the subgroups showed that an equal frequency of patients and controls had undergone endocrine therapy (0.3% in both groups, Table S3.1), either for breast cancer or for prostate cancer. The association was mainly caused by medication from the subgroups ‘antineoplastic agents’ and ‘immunosuppressants’, all of which appeared to be immunosuppressive in nature (Table S3.2). When grouped together, a significant association of immunosuppressive

**

*

Prev

alen

ce (%

)

Statin use

ALS patients

Controls

Hypercholesterolemia

Figure 3.1 Prevalence of hypercholesterolemia and statin use in ALS patients and controls * P = 0.006, calculated by logistic regression analysis for the association between hypercholesterolemia and the risk of developing ALS. ** P = 1.86 x 10-9, calculated by logistic regression analysis for the association between statin use and the risk of developing ALS.


37


drugs and a decreased risk of ALS was found (OR 0.26, 95% CI 0.08-0.86, P = 0.03). Removing current smoking and current alcohol consumption from the multivariate analyses, or adding BMI did not essentially alter the ORs (data not shown).

C9orf72 Repeat Expansion Of the 722 apparently sporadic ALS patients, 43 had a C9orf72 repeat expansion. Patients with a C9orf72 repeat expansion were significantly younger (58.5 vs. 63.3 years, P = 0.001), were more often female (58% vs. 39%, P = 0.01) and had more often a bulbar site of onset (42% vs. 32%, P = 0.21). Of the one-third (n = 732) of the controls who where analyzed, none had a C9orf72 repeat expansion. Excluding patients with a C9orf72 repeat expansion did not significantly change any of the results. To determine whether patients with a C9orf72 repeat expansion may represent a subgroup with different lifestyle and environmental factors, we subsequently analyzed the subgroup of patients with a C9orf72 repeat expansion. No significant associations were found (Table S3.3).

Table 3.2 Association of ALS with premorbid medical conditions

ALS patientsn (%)

Controlsn (%)

OR (95% CI)a P Value

Medical Condition

Cardiovascular diseases 352 (48.8) 1170 (51.6) 0.85 (0.72-1.02) 0.08

Diabetes 49 (6.8) 184 (8.1) 0.72 (0.51-1.01) 0.06

Hypercholesterolemia 188 (26.0) 702 (31.0) 0.76 (0.63-0.92) 0.006

Hypertension 233 (32.3) 784 (34.6) 0.90 (0.75-1.08) 0.26

Stroke 10 (1.4) 30 (1.3) 0.99 (0.48-2.07) 0.99

Myocardial infarction 31 (4.3) 108 (4.8) 0.90 (0.59-1.38) 0.64

Peripheral arterial disease 6 (0.8) 30 (1.3) 0.51 (0.21-1.25) 0.14

Neurodegenerative diseases 6 (0.8) 8 (0.4) 2.19 (0.74-6.50) 0.16

Parkinson's disease 3 (0.4) 5 (0.2) 1.51 (0.35-6.54) 0.58

Psychiatric disorders 16 (2.2) 44 (1.9) 1.09 (0.6-1.96) 0.78

Psychotic illness 1 (0.1) 5 (0.2) 0.67 (0.08-5.81) 0.71

Cancer 55 (7.6) 205 (9.0) 0.86 (0.62-1.18) 0.35

Infectious diseases 75 (10.4) 243 (10.7) 0.99 (0.75-1.32) 0.99

Autoimmune diseases 15 (2.1) 57 (2.5) 0.79 (0.44-1.42) 0.43

Trauma 81 (11.2) 208 (9.2) 1.36 (1.03-1.79) 0.03

Head trauma 20 (2.8) 35 (1.5) 1.95 (1.11-3.43) 0.02

Surgery 278 (38.5) 917 (40.4) 0.94 (0.79-1.12) 0.47

Abbreviations: ALS = amyotrophic lateral sclerosis; OR= odds ratio; CI = confidence interval.a ORs are adjusted for gender, age, education, current smoking and current alcohol use.


CHAPTER 3

38

Table 3.3 Prevalence of various autoimmune diseases and cancer types in ALS patients and controls

Prevalence (%)ALS patients

(n = 722)

Prevalence (%)controls

(n = 2268)

P Value

Autoimmune disease 15 (2.1) 57 (2.5) 0.51

Addison's disease 0 (0.0) 0 (0.0) -

Ankylosing spondylitis 0 (0.0) 0 (0.0) -

Autoimmune heamolytic anemia 0 (0.0) 2 (0.09) 0.43

Celiac disease 0 (0.0) 2 (0.09) 0.43

Crohn's disease 1 (0.14) 1 (0.04) 0.39

Dermatomyositis 0 (0.0) 0 (0.0) -

Graves' hyperthyroidism 1 (0.14) 5 (0.22) 0.67

Hashimoto's thyroid disease 0 (0.0) 2 (0.09) 0.43

Multiple sclerosis 1 (0.14) 0 (0.0) 0.08

Myasthenia gravis 0 (0.0) 4 (0.17) 0.26

Polymyalgia rheumatica 1 (0.14) 5 (0.22) 0.67

Psoriasis 3 (0.42) 9 (0.40) 0.95

Rheumatoid arthritis 5 (0.69) 18 (0.79) 0.79

Scleroderma 1 (0.14) 1 (0.04) 0.39

Sjogren's syndrome 0 (0.0) 2 (0.09) 0.43

Systemic lupus erythematosus 0 (0.0) 1 (0.04) 0.57

Ulcerative colitis 2 (0.28) 5 (0.22) 0.78

Vitiligo 0 (0.0) 0 (0.0) -

Cancer 55 (7.6) 205 (9.0) 0.24

Breast cancer 10 (1.4) 37 (1.6) 0.64

Colorectal cancer 4 (0.6) 22 (1.0) 0.29

Non-Hodgkin’s Lymphoma 1 (0.1) 5 (0.2) 0.67

Kidney cancer 0 (0.0) 9 (0.4) -

Leukemia (all) 1 (0.1) 5 (0.2) 0.67

Lung cancer 2 (0.3) 4 (0.2) 0.60

Ovarian cancer 0 (0.0) 3 (0.1) -

Prostate cancer 9 (1.2) 32 (1.4) 0.74

Skin cancer 17 (2.4) 57 (2.5) 0.81

Urinary bladder cancer 1 (0.1) 8 (0.4) 0.36

Uterine corpus and cervical cancer 5 (0.7) 11 (0.5) 0.51

Other types of cancera 5 (0.7) 12 (0.5) 0.61a Other types of cancer were cancer from the pancreas, liver, spleen, thyroid, brain, vocal cord, salivary gland, and testicles.


39


Survival and Age at Onset Analyses None of the categories of medical conditions or premorbid use of medication was associated with survival. Specifically, patients with hypercholesterolemia or patients using statins did not have a significant difference in survival compared to patients without hypercholesterolemia or not using statins (HR 1.14, 95% CI 0.92-1.41, P = 0.22; and HR 1.19, 95% CI 0.89-1.60, P = 0.23 respectively). Cox regression analysis of age at onset of ALS patients showed that diabetes, myocardial infarction and premorbid use of medication of three ATC main groups (alimentary tract

Table 3.4 Association of ALS with premorbid use of medication, based on the ATC classification system

ALS patientsn (%)

Controlsn (%)


(A) Alimentary tract and metabolism

130 (18.0) 408 (18.0) 0.92 (0.73-1.16) 0.50

(B) Blood and blood-forming organs

118 (16.3) 381 (16.8) 0.93 (0.73-1.19) 0.56

(C) Cardiovascular system 251 (34.8) 887 (39.1) 0.79 (0.65-0.96) 0.02

(D) Dermatologicals 1 (0.1) 12 (0.5) 0.26 (0.03-2.05) 0.20

(G) Genito-urinary system and sex hormones

40 (5.5) 134 (5.9) 0.96 (0.67-1.40) 0.85

(H) Systemic hormonal preparations

23 (3.2) 104 (4.6) 0.69 (0.43-1.12) 0.14

(J) Anti-infectives for systemic use 2 (0.3) 15 (0.7) 0.51 (0.12-2.28) 0.38

(L) Antineoplastic and immunomodulating agents

6 (0.8) 40 (1.8) 0.35 (0.14-0.91) 0.03

(M) Musculo-skeletal system 49 (6.8) 152 (6.7) 0.99 (0.70-1.40) 0.94

(N) Nervous system 106 (14.7) 240 (10.6) 1.29 (1.00-1.69) 0.05

(P) Antiparasitic products, insectides and repellents

0 (0) 5 (0.2) - -

(R) Respiratory system 63 (8.7) 153 (6.7) 1.23 (0.89-1.69) 0.21

(S) Sensory organs 6 (0.8) 26 (1.1) 0.82 (0.33-2.02) 0.66

Abbreviations: ALS = amyotrophic lateral sclerosis; ATC = anatomical therapeutic chemical; OR = odds ratio; CI = confidence interval. a ORs are adjusted for gender, age, education, current smoking and current alcohol use.


CHAPTER 3

40

and metabolism; blood and blood forming organs; and cardiovascular system) were significantly related to age at onset. However, because of a potentially confounding factor – age (older individuals are more likely to have comorbidities), the analysis had to be adjusted appropriately for this effect, in two different ways. First of all, an interaction term of case-control status with medical history or premorbid use of medication was introduced into the model. The interaction terms for diabetes, myocardial infarction and for the three medication groups were non-significant. Secondly, the multivariate Cox regression analysis was performed in controls using the filling out of the questionnaire as the event. This showed a similar association for these variables with ‘age at event’ in controls compared to the association with ‘age at onset’ in patients. Both indicate that the relation between medical history, premorbid use of medication and age at onset, is an age-related effect and not an ALS-specific effect.

DISCUSSION

In this large population-based, case-control study, we found that hypercholesterolemia and statin use prior to symptom onset are associated with a decreased risk of ALS, which indicates a premorbid favorable lipid profile in at least a subset of ALS patients. Secondly, the use of immunosuppressive drugs is significantly lower in ALS patients, which could suggest a protective effect on ALS susceptibility. Thirdly, antecedent head trauma may be a risk factor for ALS. Lastly, we did not find evidence for an effect of prior medical conditions or medications on survival, symptom onset or C9orf72 genotype.Hypercholesterolemia is less frequently reported among ALS patients compared to controls, which is a proxy for a favorable lipid profile. This is in line with our previous findings in a smaller, independent, hospital-based study that ALS patients have lower blood lipid levels and use less statins compared to controls.2 In contrast, in a French ALS cohort, the frequency of hyperlipidemia was higher in patients compared to controls,19 while in an Italian study no altered lipid levels were found in ALS patients.20 These discrepancies are probably due to differences in the control population, or they might reflect differences between populations and lifestyles (i.e. differences in dietary habits). The observation of a favorable lipid profile is consistent with the significantly lower premorbid BMI found in ALS patients compared to controls (24.1 versus 25.6 kg/m2, P < 0.001, shown in Table 3.1) as was also reported in previous studies.2, 21 Adding BMI to the multivariate model resulted in a non-significant association between hypercholesterolemia and ALS, which suggests that BMI is either a confounder or that hypercholesterolemia and BMI are in the same cascade. In contrast with lower lipid levels and a lower BMI, there is evidence for a high fat, high caloric dietary intake in ALS patients before onset of symptoms.22 This discrepancy in energy homeostasis can be explained by an increased metabolic rate, which may be present in ALS patients prior to symptom onset as has also been demonstrated in SOD1 mouse models of ALS.23, 24


41


In line with a lower frequency of premorbid hypercholesterolemia, we found a lower use of statins in patients, which is in contrast with previous studies from the WHO Foundation Collaborating Centre for International Drug Monitoring and the Food and Drug Administration. Both registered disproportionally high prevalence of ALS cases among subjects using statins.25, 26 This signal could not, however, be validated by retrospective pooling of 41 clinical trials on statin use or within a population-based, case-control study in Denmark.26, 27 Our study even showed evidence for a possible protective effect of statins in ALS. A neuroprotective and anti-inflammatory effect of statins has been reported in other neurodegenerative diseases (i.e. Alzheimer’s and Parkinson’s disease),28, 29 in multiple sclerosis,30 as well as in the wobbler mouse model of motor neuron degeneration.31 We did not find a longer survival in patients with hypercholesterolemia or a shorter survival in patients on statins as was reported in case-control studies in ALS.2, 19, 32, 33 Most of those studies, however, found an association with survival in a univariate model without adjusting for covariates. Two studies showed that the association disappeared after adjustment for age at onset, site of onset, BMI or forced vital capacity.2, 32, 34 This was also true in our study: in a univariate model, there was a significant association between statin use and a shorter survival (HR 1.46, 95% CI 1.09-1.94, P = 0.01). The difference with our null finding in the multivariate model can be explained, since patients using statins were significantly older than patients not using statins (median 67 vs. 61 years, P = 0.002). In addition, in the multivariate model both age at onset and site of onset were significantly associated with survival (in contrast with statin use), implicating that these factors influence disease progression.We found a significant association for trauma, specifically head trauma, and ALS risk. Head trauma consisted of severe traumatic brain injuries (i.e. skull fractures, concussion or intracranial hemorrhage); no microtrauma were included in our study. Previously, head trauma was considered to be a risk factor for ALS, initially in a study on Italian soccer players,35 followed by other reports in veterans and population studies.12, 36, 37 Although more research is required to confirm the association,38 our study further supports head trauma as a risk factor for ALS. Even when analyzing head trauma occurring at least five years prior to symptom onset, to exclude possible injuries due to incipient ALS, an association was found with ALS (OR 1.86, 95% CI 1.01-3.42, P < 0.05). The use of immunosuppressive agents is significantly lower in ALS patients, which may suggest that immunosuppressive agents have a protective effect on ALS susceptibility in controls, but the relatively low frequency of their use makes it difficult to determine this with certainty. A lower use of immunosuppressive agents is at least consistent with the lack of association with autoimmune diseases in our population. In contrast to our study, two hospital-based, record linkage studies found an increased frequency of autoimmune diseases amongst ALS patients and their relatives.9, 10 Furthermore, cardiovascular diseases, psychiatric disorders and cancer, all of which have previously been linked to ALS,2-8 did not show an association with the risk of developing ALS in our study. These differences may be due to a population-specific effect, with environmental and genetic diversity between populations influencing phenotype and predisposition.


CHAPTER 3

42

A limitation of the study is that by not applying a multiple test adjustment method, we might have rejected a true null hypothesis. Due to the exploratory nature of this study, the use of multiple test adjustment can be considered too conservative.39 Furthermore, we acknowledge that in questionnaires, differential recall of exposures between patients and controls may be a limitation.40 Patients may be more eager to find an explanation for their disease, which is why they tend to over-report events. In addition, cognitive and executive impairment, which occurs in ALS, might have influenced recall as well. This risk of recall bias was reduced by calling participants to complete the questionnaires and by blinding both the participants, and the interviewers for the hypotheses being tested. For validation of our questionnaire, we checked the medical records of 122 participants, which showed a ≥95% concordance between the questionnaires and the medical records. In conclusion, the diagnosis of ALS may be associated with a favorable lipid profile prior to symptom onset in at least a subpopulation of ALS. Furthermore, trauma, and specifically head trauma, is shown to be a risk factor for ALS. These findings provide new insight into possible pathogenic mechanisms in ALS.

Ethical standardEthical approval was obtained from the institutional review board of the University Medical Center Utrecht. All participants gave written informed consent for inclusion in the study.


43


REFERENCES

1. Al-Chalabi A and Hardiman O. The epidemiology of ALS: a conspiracy of genes, environment and time. Nat Rev Neurol. 2013; 9: 617-28.

2. Sutedja NA, et al. Beneficial vascular risk profile is associated with amyotrophic lateralsclerosis. J Neurol Neurosurg Psychiatry. 2011; 82: 638-42.

3. TurnerMR, et al. Cardiovascular fitness as a risk factor for amyotrophic lateral sclerosis:indirect evidence from record linkage study. J Neurol Neurosurg Psychiatry. 2012; 83: 395-8.

4. Valavanis A, et al. Amyotrophic lateral sclerosis after embolization of cerebral arterioveneous malformations. J Neurol. 2014; 261: 732-7.

5. Byrne S, et al. Aggregation of neurologic and neuropsychiatric disease in amyotrophic lateral sclerosis kindreds: a population-based case-control cohort study of familial and sporadic amyotrophic lateral sclerosis. Ann Neurol. 2013; 74: 699-708.

6. Schreiber H, et al. Cognitive function in bulbar- and spinal-onset amyotrophic lateral sclerosis. A longitudinal study in 52 patients. J Neurol. 2005; 252: 772-81.

7. Freedman DM, et al. The association between cancer and amyotrophic lateral sclerosis. Cancer Causes Control. 2013; 24: 55-60.

8. Fois AF, et al. Cancer in patients with motor neuron disease, multiple sclerosis and Parkinson's disease: record linkage studies. J Neurol Neurosurg Psychiatry. 2010; 81: 215-21.

9. Turner MR, et al. Autoimmune disease preceding amyotrophic lateral sclerosis: an epidemiologic study. Neurology. 2013; 81: 1222-5.

10. Hemminki K, et al. Familial risks for amyotrophic lateral sclerosis and autoimmune diseases. Neurogenetics. 2009; 10: 111-6.

11. Beghi E, et al. Amyotrophic lateral sclerosis, physical exercise, trauma and sports: results of a population-based pilot case-control study. Amyotroph Lateral Scler. 2010; 11: 289-92.

12. Pupillo E, et al. Trauma and amyotrophic lateral sclerosis: a case-control study from a population-based registry. Eur J Neurol. 2012; 19: 1509-17.

13. Statistics Netherlands. Population. 2011.14. Brooks BR, et al. El Escorial revisited: revised criteria for the diagnosis of amyotrophic lateral

sclerosis. Amyotroph Lateral Scler Other Motor Neuron Disord. 2000; 1: 293-9.15. DeJesus-Hernandez M, et al. Expanded GGGGCC hexanucleotide repeat in noncoding region

of C9orf72 causes chromosome 9p-linked FTD and ALS. Neuron. 2011; 72: 245-56.16. van Rheenen W, et al. Hexanucleotide repeat expansions in C9orf72 in the spectrum of motor

neuron diseases. Neurology. 2012; 79: 878-82.17. WorldHealthOrganization.GuidelinesforATCclassificationandDDDassignment.2012.18. Logroscino G, et al. Incidence of amyotrophic lateral sclerosis in Europe. J Neurol Neurosurg

Psychiatry. 2010; 81: 385-90.19. Dupuis L, et al. Dyslipidemia is a protective factor in amyotrophic lateral sclerosis. Neurology.

2008; 70: 1004-9.20. Chio A, et al. Lower serum lipid levels are related to respiratory impairment in patients with

ALS. Neurology. 2009; 73: 1681-5.21. O'Reilly EJ, et al. Premorbid body mass index and risk of amyotrophic lateral sclerosis.

Amyotroph Lateral Scler Frontotemporal Degener. 2013; 14: 205-11.22. Nelson LM, et al. Population-based case-control study of amyotrophic lateral sclerosis in

western Washington State. II. Diet. Am J Epidemiol. 2000; 151: 164-73.23. Bouteloup C, et al. Hypermetabolism in ALS patients: an early and persistent phenomenon. J

Neurol. 2009; 256: 1236-42.24. Dupuis L, et al. Evidence for defective energy homeostasis in amyotrophic lateral sclerosis:

benefitofahigh-energydietinatransgenicmousemodel.ProcNatlAcadSciUSA.2004;101: 11159-64.

25. Edwards IR, et al. Statins, neuromuscular degenerative disease and an amyotrophic lateral sclerosis-like syndrome: an analysis of individual case safety reports from vigibase. Drug Saf. 2007; 30: 515-25.


CHAPTER 3

44

26. Colman E, et al. An evaluation of a data mining signal for amyotrophic lateral sclerosis and statins detected in FDA's spontaneous adverse event reporting system. Pharmacoepidemiol Drug Saf. 2008; 17: 1068-76.

27. Sorensen HT, et al. Statin use and risk of amyotrophic lateral sclerosis and other motor neuron disorders. Circ Cardiovasc Qual Outcomes. 2010; 3: 413-7.

28. Arvanitakis Z, et al. Statins, incident Alzheimer disease, change in cognitive function, and neuropathology. Neurology. 2008; 70: 1795-802.

29. Wahner AD, et al. Statin use and the risk of Parkinson disease. Neurology. 2008; 70: 1418-22.30. Ciurleo R, et al. Role of statins in the treatment of multiple sclerosis. Pharmacol Res. 2014; 87:

133-43.31. Iwamoto K, et al. Atorvastatin treatment attenuates motor neuron degeneration in wobbler

mice. Amyotroph Lateral Scler. 2009; 10: 405-9.32. DroryVE,etal.Influenceofstatinstreatmentonsurvivalinpatientswithamyotrophiclateral

sclerosis. J Neurol Sci. 2008; 273: 81-3.33. Dorst J, et al. Patients with elevated triglyceride and cholesterol serum levels have a prolonged

survival in amyotrophic lateral sclerosis. J Neurol. 2011; 258: 613-7.34. Zinman L, et al. Are statin medications safe in patients with ALS? Amyotroph Lateral Scler.

2008; 9: 223-8.35. Chio A, et al. Severely increased risk of amyotrophic lateral sclerosis among Italian professional

football players. Brain. 2005; 128: 472-6.36. Schmidt S, et al. Association of ALS with head injury, cigarette smoking and APOE genotypes.

J Neurol Sci. 2010; 291: 22-9.37. Chen H, et al. Head injury and amyotrophic lateral sclerosis. Am J Epidemiol. 2007; 166: 810-

6.38. Armon C and Nelson LM. Is head trauma a risk factor for amyotrophic lateral sclerosis? An

evidence based review. Amyotroph Lateral Scler. 2012; 13: 351-6.39. Bender R and Lange S. Adjusting for multiple testing--when and how? J Clin Epidemiol. 2001;

54: 343-9.40. Coughlin SS. Recall bias in epidemiologic studies. J Clin Epidemiol. 1990; 43: 87-91.


45



Table S3.1 Association of ALS with medication use of the ATC coded subgroups of the “cardiovascular system”, and “antineoplastic and immunomodulating agents”

ATCcode

ALS patientsn (%)

Controlsn (%)


Subgroups

Cardiovascular systemb

Cardiac therapy C01 21 (2.9) 70 (3.1) 0.95 (0.57-1.59) .85

Antihypertensives C02 3 (0.4) 19 (0.8) 0.33 (0.08-1.45) .14

Diuretics C03 69 (9.6) 210 (9.3) 1.01 (0.74-1.36) .97

Peripheral vasodilators C04 0 (0.0) 1 (0.0) - -

Vasoprotectives C05 1 (0.1) 0 (0.0) - -

Beta blocking agents C07 108 (15.0) 365 (16.1) 0.89 (0.70-1.14) .36

Calcium channel blockers C08 45 (6.2) 161 (7.1) 0.88 (0.62-1.25) .47

Renin-angiotensin system C09 128 (17.7) 415 (18.3) 0.96 (0.76-1.21) .72

Lipid modifying agents C10 90 (12.5) 487 (21.5) 0.45 (0.35-0.59) <.001

Antineoplastic and immunomodulating agents

Antineoplastic agents L01 0 (0.0) 6 (0.3) - -

Endocrine therapy L02 2 (0.3) 7 (0.3) 0.80 (0.16-4.06) .79

Immunostimulants L03 0 (0.0) 0 (0.0) - -

Immunosuppressants L04 4 (0.6) 27 (1.2) 0.31 (0.09-1.04) .06

Abbreviations: ATC = anatomical therapeutic chemical; ALS = amyotrophic lateral sclerosis; OR = odds ratio; CI = confidence interval. a ORs are adjusted for gender, age, education, current smoking and current alcohol consumption. b ATC subgroup C06 does no longer exist.


CHAPTER 3

46

Table S3.2 Prevalences of medication use of the ATC coded subgroups of “antineoplastic and immunomodulating agents” in ALS patients and controls

ATC code ALS patientsn (%)

Controlsn (%)

Subgroups

Antineoplastic agents L01 0 (0.0) 6 (0.3)

Chlorambucil L01AA02 0 (0.0) 1 (0.0)

Methotrexate L01BA01 0 (0.0) 1 (0.0)

Imatinib L01XE01 0 (0.0) 1 (0.0)

Hydroxycarbamide L01XX05 0 (0.0) 2 (0.1)

Chemotherapy (not specified) L01 0 (0.0) 1 (0.0)

Endocrine therapya L02 2 (0.3) 7 (0.3)

Goserelin L02AE03 0 (0.0) 1 (0.0)

Tamoxifen L02BA01 1 (0.1) 2 (0.1)

Bicalutamide L02BB03 0 (0.0) 4 (0.2)

Letrozole L02BG04 1 (0.1) 1 (0.0)

Immunostimulants L03 0 (0.0) 0 (0.0)

Immunosuppressantsb L04 4 (0.6) 27 (1.2)

Leflunomide L04AA13 0 (0.0) 1 (0.0)

Etanercept L04AB01 0 (0.0) 3 (0.1)

Ciclosporin L04AD01 1 (0.1) 2 (0.1)

Tacrolimus L04AD02 0 (0.0) 2 (0.1)

Azathioprine L04AX01 2 (0.3) 4 (0.2)

Thalidomide L04AX02 0 (0.0) 1 (0.0)

Methotrexate L04AX03 1 (0.1) 17 (0.7)

Abbreviations: ATC = anatomical therapeutic chemical; ALS = amyotrophic lateral sclerosis. a Total number of controls using endocrine therapy (n = 7) is less than the combined individual numbers, because 1 control used >1 endocrine therapy. b Total number of controls using immunosuppressant (n = 27) is less than the combined individual numbers, because three controls used >1 immunosuppressant.


47


Table S3.3 Association of ALS with premorbid medical conditions, in sporadic ALS patients with a C9orf72 repeat expansion (n = 43) compared to 762 controls with no C9orf72 repeat expansion

ALS patientsn (%)

Controlsn (%)


Cardiovascular diseases 18 (41.9) 394 (51.7) 0.81 (0.43-1.56) .54

Diabetes 0 (0.0) 62 (8.1) - -

Hypercholesterolemia 9 (20.9) 235 (30.8) 0.70 (0.33-1.52) .37

Hypertension 13 (30.2) 265 (34.8) 0.94 (0.47-1.88) .87

Stroke 0 (0.0) 10 (1.3) - -

Myocardial infarction 0 (0.0) 40 (5.2) - -

Peripheral arterial disease 0 (0.0) 7 (0.9) - -

Neurodegenerative diseases 0 (0.0) 2 (0.3) - -

Parkinson's disease 0 (0.0) 0 (0.0) - -

Psychiatric disorders 1 (2.3) 14 (1.8) 0.96 (0.12-7.66) .97

Psychotic illness 0 (0.0) 4 (0.5) - -

Cancer 1 (2.3) 68 (8.9) 0.28 (0.04-2.10) .22

Infectious diseases 4 (9.3) 80 (10.5) 0.89 (0.31-2.59) .83

Autoimmune diseases 1 (2.3) 22 (2.9) 0.57 (0.07-4.58) .60

Trauma 2 (4.7) 79 (10.4) 0.47 (0.11-1.98) .30

Head trauma 0 (0.0) 11 (1.4) - -

Surgery 15 (34.9) 304 (39.9) 0.81 (0.42-1.55) .52

Abbreviations: ALS = amyotrophic lateral sclerosis; OR= odds ratio; CI = confidence interval. a ORs are adjusted for gender, age, education, current smoking and current alcohol use.



Mark H B Huisman1, Meinie Seelen1, Perry T C van Doormaal1, Sonja W de Jong1, Jeanne H M de Vries2, Anneke J van der Kooi3, Marianne de Visser3,

H Jurgen Schelhaas4, Leonard H van den Berg1*, Jan H Veldink1*

1 Department of Neurology, Brain Center Rudolf Magnus, University Medical Center Utrecht, The Netherlands.

2 Division of Human Nutrition, Wageningen University, the Netherlands. 3 Department of Neurology, Amsterdam Medical Center,

University of Amsterdam, The Netherlands. 4 Department of Neurology, Donders Institute for Brain, Cognition and Behaviour,

Center for Neuroscience, Radboud University Nijmegen Medical Center, The Netherlands.

*Joint last authors

In preparation

Pre-symptomatic BMI, dietary fat

and alcohol consumption as independent risk factors

for amyotrophic lateral sclerosis

CHAPTER 4


CHAPTER 4

50

ABSTRACT

BackgroundDietary intake may influence pathophysiological mechanisms in sporadic amyotrophic lateral sclerosis (ALS). We, therefore, aimed to systematically determine the relation between premorbid dietary intake and the risk of sporadic ALS, in order to provide better insight into which mechanisms are possibly involved in ALS pathophysiology.

MethodsIn a population-based case-control study in The Netherlands, including 674 patients and 2,093 controls, we studied the premorbid intake of nutrients in relation to the risk of ALS, using a food frequency questionnaire adjusted for confounding factors and corrected for multiple comparisons, while minimizing recall bias.

FindingsPre-symptomatic total daily energy intake in patients was significantly higher compared with controls (p < 0·01), while pre-symptomatic BMI was significantly lower in patients (p = 0·02). Higher premorbid intakes of total fat, saturated fat, trans fatty acids, and cholesterol were associated with an increased risk of ALS, while higher intake of alcohol was associated with a decreased risk. These associations were independent of total energy intake, age, gender, BMI, education, smoking and lifetime physical activity.

InterpretationThe combination of a positive association of a low premorbid BMI and a high fat intake, together with prior evidence from ALS SOD1 mouse models and earlier reports on premorbid BMI, supports a role for altered energy metabolism prior to clinical onset of ALS.


51

Premorbid BMI, dietary intake and ALS risk

INTRODUCTION

Amyotrophic lateral sclerosis (ALS) is a devastating neurodegenerative disorder, characterized by the progressive loss of upper and lower motor neurones, with a median survival from onset of three years, no cure and an aetiology that is still poorly understood.1,

2 Oxidative stress, mitochondrial dysfunction, excitotoxicity, disrupted axonal transport and inflammation are mechanisms that are potentially involved in ALS pathophysiology. Some of these mechanisms may be influenced by dietary nutrients. The intake of dietary antioxidants, for example, may reduce oxidative stress.3, 4

Previous studies did not identify a consistent nutrient that modifies susceptibility to ALS.3-13 Most associations have not been replicated, and contradictory results exist for the association with total fat intake.3, 7 Decreased risk of ALS with higher levels of vitamin E intake, a potent cellular antioxidant, is one of the associations that has been reported more than once.4, 6, 11 Furthermore, a recent study replicated the observation that a higher intake of polyunsaturated fatty acids (PUFA) is associated with a decreased risk of ALS.4,

14 These studies suggest that nutrients, that can have direct neuroprotective properties or influence pathways known to be involved in ALS pathogenesis, may be associated with ALS. Since diet is highly modifiable, in our large population-based case-control study, we set out to test the relation between premorbid intake of many nutrients and the risk of ALS and the progression of the disease, adjusted for confounding factors and corrected for multiple comparisons.

METHODS

Study design and participantsThe Prospective ALS study The Netherlands (PAN) is a population-based case-control study performed in The Netherlands from January 1st, 2006, until September 30st, 2011. During the study period, all patients newly diagnosed with possible, probable (laboratory-supported) or definite ALS, according to the revised El Escorial criteria, were included.15 Multiple sources were used to ensure complete case ascertainment: neurologists, rehabilitation physicians, the Dutch Neuromuscular Patient Association and our ALS website. Medical records were scrutinized for eligibility of the patients, excluding patients with an ALS-mimic syndrome or with a first, second or third degree family member with ALS, defined as familial ALS. Patients with a C9orf72 repeat expansion were excluded from our analysis. An expanded repeat in C9orf72 was assessed by performing a repeat-primed PCR reaction as described previously.16 Population-based controls were selected from the register of the general practitioner (GP) taking care of the participating patient with ALS. In The Netherlands, the health care system ensures that every inhabitant is registered at a general practitioner which makes this record representative of the population. The GP was asked to select individuals from


CHAPTER 4

52

his register in alphabetical order, starting at the surname of the patient. The controls were matched to the patients for gender and age (plus or minus five years). Blood-relatives or spouses of the patients were not eligible to be controls to prevent overmatching. Ethical approval was provided by the institutional review board of the University Medical Centre Utrecht. All participants gave written informed consent.

ProceduresPatients and controls were asked to fill in a 199-item food frequency questionnaire (FFQ) that covered the food consumption over the previous month. However, if dietary habits had changed since onset of symptoms, patients were asked to recall their dietary habits over the one-month period prior to the onset of muscle weakness or bulbar signs, to avoid a possible influence of disease on their intake. Food items for the original questionnaire were chosen on the basis of data from the Dutch National Food Consumption Survey of 1992,17 and updated based on a 1998 survey.18 The selected food items for this FFQ covered about 95% of the intake of total energy, total fat, fatty acids and cholesterol of the Dutch population and was validated for this purpose.19 Considering the hypotheses of the present study, the questionnaire was extended with questions on the intake of foods which contributed > 0·5% to the population intake of protein, carbohydrates, dietary fibres, alcohol, calcium, vitamin B2, vitamin C, vitamin E, lycopene, flavonoids, glutamate and phyto-oestrogens. For several food items, additional questions were included on preparation method or portion sizes. Consumed amounts were calculated using standard household measures.20 For nutrient calculations, the 2006 Dutch Food Composition Table was used for energy, macronutrients and vitamin C;21 national reports by TNO Nutrition and Food Research for calcium, vitamin B2 and vitamin E; publications for flavonoids;22,

23 the US Department of Agriculture table for phyto-oestrogens (isoflavones);24 publications for glutamate and monosodium glutamate;25-29 and the US Department of Agriculture table for lycopene.24

If necessary, patients and control subjects were contacted by telephone to clarify inconsistencies or missing data in the questionnaire. FFQs of 5 controls were not included in the analyses because of implausible low or high reported energy intakes. For this, theoretical physical activity levels (PALs) were calculated, dividing reported energy intake by the Basal Metabolic Rate, using Schofield’s formulae, and compared to the lower and upper cut-off limits for these PALs.30 All questionnaires remained anonymous during the analyses, and all data were entered in a blinded fashion.A second, self-administered general questionnaire was filled out by the participants to obtain data on age, gender, level of education, smoking habits, anthropometrical characteristics and a lifetime history of occupations, sports and hobbies.31 Data on survival of patients, up to February 1, 2012, was monitored using the municipal population register.


53


Statistical analysisBaseline characteristics were tested for differences using Pearson’s Chi square and the Mann-Whitney U test. To determine odds ratios (ORs) for the association between the intake of a specific nutrient and ALS, we performed a binary logistic regression with three levels of adjustment: (1) adjusting for age (at onset for patients, age at which questionnaire was completed in controls), gender and education; (2) additionally adjusted for BMI (premorbid in patients), smoking (current or not) and lifetime physical activity; and (3) additionally adjusted for total energy intake. The lifetime physical activity was calculated from the lifetime history of occupations, sports and hobbies, and has been described elsewhere.31 We also determined the relation between nutrient intake and risk of ALS using the multivariate nutrient density model designed by Willett et al, which is another frequently used model, to account for total energy intake:32

The meaning of the coefficient β1 for the nutrient density (i.e. energy provided by nutrient / total energy) is the difference in disease risk associated with a difference in 1% of energy from the nutrient while total energy intake is kept constant. In nutrients that do not yield energy, nutrient density was expressed as nutrient intake in milligrams per 1,000 kilocalories of energy intake. In this analysis we also adjusted for age, gender, education, BMI, smoking and physical activity.An additional logistic regression, with the same covariates, was performed in which nutrient intake was categorized into quintiles, based on the nutrient intake in controls. The lowest quintile served as the reference group and the five-level variables were also entered into the model as continuous variables to determine whether there was a linear trend. Those nutrients that were significantly associated with ALS, either in the analysis with absolute values of nutrient intake or in the analysis with quintiles of intake, were analyzed together in a multivariate binary logistic regression to determine which of these nutrients were independently associated with ALS. This analysis was performed with the maximal level of adjustment.Finally, Cox regression analysis was performed to determine the association between survival from onset and the intake nutrients. The hazard ratios (HR) derived from these analyses were adjusted for gender, age at onset, site of onset, premorbid BMI, energy intake, education, current smoking, and lifetime physical activity.31 The same method was used to determine the effect of nutrient intake on the age at onset of ALS patients. To adjust appropriately for age, an interaction term of diagnosis and nutrient or dietary pattern was introduced to the Cox regression analysis using age at time of completing the questionnaire for controls. All tests were two-sided, and a Bonferroni correction was applied to the alpha level to adjust for multiple comparisons. The Bonferroni adjusted p-values are shown in the tables.


CHAPTER 4

54

Role of the funding sourceThe funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. All authors had full access to all the data in the study and the corresponding author had final responsibility for the decision to submit for publication.

RESULTS

Informed consent to participate in the study was given by 885 (90%) of a total of 986 eligible patients identified between January 1, 2006 and September 31, 2011. Of the questionnaires sent to these 885 patients, 747 were returned (84%). 674 (87%) of these patients had completed the questionnaires without any omissions and were included in the analyses. A total of 2480 population-based controls were selected from the GP’s register, and 2385 of these returned their questionnaire (response rate 96%). Of these 2385 controls, 2093 (88%) had completed the questionnaires without any missing value and were included. Table 4.1 shows the characteristics of the 674 patients and 2093 controls included in the analyses. Gender, mean age at onset, and frequency of bulbar onset did not differ significantly between responders and non-responders. Cases and controls were similar for the matching variables, gender and age. Pre-symptomatic BMI was significantly lower in patients than controls (p=0·02) (Table 4.1). In contrast, pre-symptomatic daily energy intake as calculated from the FFQ was significantly higher in patients compared with controls (p<0·01). Total lifetime physical activity scores (during both leisure and occupational time) did not differ between patients and controls (p=0·2). Table 4.2 shows the adjusted ORs for the association between the premorbid intake of individual nutrients and the risk of ALS. Higher intakes of total fat, saturated fat, trans fatty acids, and cholesterol were independently associated with an increased risk of ALS, irrespective of the level of adjustment and irrespective of using absolute intake or nutrient density in the analysis. In the maximal adjusted model, higher intakes of vegetable protein, polysaccharides, fibres, and flavonoids were associated with a decreased risk of ALS. The relation with quintiles of intake of these nutrients, and the p values for trend across quintiles are illustrated in Figure 4.1 (significant associations) and in Supplementary Figure S4.1 (non-significant associations). Figure 4.1 shows that alcohol is significantly related to a decreased risk of ALS (p for trend: <0·001). From the different dietary fats, we only included the intake of saturated fat in the multivariate analysis, since both trans fatty acids and cholesterol were highly correlated with the intake of saturated fat (trans fatty acids: r = 0·95; cholesterol: r = 0·73). Besides saturated fat, the intake of vegetable protein, polysaccharides, fibres, alcohol and flavonoids were analyzed together in the multivariate model, since these nutrients were significantly associated with risk of ALS in the maximal adjusted model or in the analysis with quintiles of intake (Table 4.2, Figure 4.1). The multivariate analysis shows that only a higher intake


55



Variable ALS patients Controls P value

(n=674) (n=2,093)

Age at first weakness, yr, mean (SD)a 62·4 (11·0) 62·6 (10·0) 0·9

Age at diagnosis, yr, mean (SD) 63·6 (11·0)

Gender: M, n (%) 418 (62) 1219 (58) 0·1

Bulbar onset, n (%) 218 (32)


Definite 119 (18)

Probable 301 (45)

Probable lab supported 126 (19)

Possible 128 (19)

Education, n (%)

No education / primary school 60 (9) 128 (6) 0·01

Secondary school 448 (67) 1369 (66)

College / University 166 (25) 594 (28)

Current smoking, n (%) 133 (20) 277 (13) <0·01

Lifetime physical activity, activity score, median (IQR) 3·8 (2·0-6·1) 3·6 (2·1-5·6) 0·2

Body mass index, kg/m2, mean (SD) 25·7 (4·0) 26·0 (3·7) 0·02

Energy intake, kcal/day, mean (SD) 2,258 (730) 2,119 (619) <0·01a Age at onset in patients, and age at which the questionnaire was completed in controls. ALS = amyotrophic lateral sclerosis; IQR = interquartile range; SD = standard deviation

Figure 4.1 Odds ratios for the relationship between ALS and quintiles of nutrient intake. Adjusted for energy intake, age (at onset in patients; at questionnaire in controls), gender, BMI, education, current smoking, and lifetime physical activity. P values shown are for the trend across quintiles.


CHAPTER 4

56

Tab

le 4

.2 A

djus

ted

OR

s fo

r th

e re

lati

onsh

ip b

etw

een

AL

S an

d nu

trie

nt in

take

Bin

ary

logi

stic

reg

ress

ion

wit

h a

bsol

ute

nu

trie

nt

inta

ke

Nu

trie

nt

den

sity

mod

el

Nu

trie

nt

Adj

ust

ed O

Ra

(9

5% C

I)p

val

ueb

Adj

ust

ed O

Rc

(9

5% C

I)p

val

ueb

Adj

ust

ed O

Rd

(9

5% C

I)p

val

ueb

Adj

ust

ed O

Rd

(9

5% C

I)p

val

ueb

Pro

tein

, tot

al1·

18 (1

·06-

1·31

)0·

002

1·19

(1·0

7-1·

33)

0·00

10·

97 (0

·80-

1·18

)0·

8

0·99

(0·9

5-1·

04)

0·7

V

eget

able

1·03

(0·8

3-1·

27)

0·8

1·03

(0·8

3-1·

28)

0·8

0·43

(0·3

0-0·

61)

<0·0

01

0·85

(0·7

8-0·

91)

<0·0

01

A

nim

al1·

37 (1

·18-

1·58

)<0

·001

1·39

(1·2

0-1·

61)

<0·0

011·

25 (1

·03-

1·52

)0·

02

1·04

(1·0

0-1·

09)

0·05

Fat,

tota

l1·

09 (1

·05-

1·12

)<0

·001

1·08

(1·0

5-1·

11)

<0·0

011·

14 (1

·07-

1·23

)<0

·001

1·

03 (1

·01-

1·05

)<0

·001

Sa

tura

ted

1·27

(1·1

8-1·

36)

<0·0

011·

26 (1

·17-

1·35

)<0

·001

1·43

(1·2

5-1·

64)

<0·0

01

1·09

(1·0

6-1·

13)

<0·0

01

M

onou

nsat

urat

ed1·

23 (1

·13-

1·33

)<0

·001

1·22

(1·1

2-1·

32)

<0·0

011·

24 (1

·04-

1·47

)0·

02

1·05

(1·0

1-1·

09)

0.02

Po

lyun

satu

rate

d1·

19 (1

·05-

1·35

)0·

007

1·18

(1·0

4-1·

34)

0·01

0·92

(0·7

6-1·

13)

0·4

0·

97 (0

·92-

1·01

)0.

2

Alp

ha L

ipol

ic A

cid

(AL

A)

1·02

(1·0

1-1·

04)

0·00

71·

02 (1

·01-

1·04

)0·

008

1·00

(0·9

8-1·

02)

0·9

0·

92 (0

·53-

1·59

)0.

8

Eic

osap

enta

enoi

c ac

id (E

PA)

1·06

(0·9

2-1·

22)

0·4

1·06

(0·9

2-1·

23)

0·4

1·03

(0·8

9-1·

19)

0·7

1·

01 (0

·05-

21·0

)0.

99

Doc

osah

exae

noic

aci

d (D

HA

)1·

06 (0

·95-

1·18

)0·

31·

06 (0

·96-

1·18

)0·

31·

03 (0

·93-

1·15

0·6

1·

21 (0

·13-

11·2

)0.

9

Om

ega

3 fa

tty

acid

s, to

tal

1·02

(1·0

1-1·

04)

0·00

61·

02 (1

·01-

1·04

)0·

006

1·00

(0·9

8-1·

02)

0·9

0·

94 (0

·57-

1·55

)0.

8

Tra

ns fa

tty

acid

s1·

03 (1

·02-

1·04

)<0

·001

1·03

(1·0

2-1·

04)

<0·0

011·

03 (1

·01-

1·05

)0·

001

2·

19 (1

·42-

3·38

)<0

·001

C

hole

ster

ol1·

45 (1

·31-

1·61

)<0

·001

1·46

(1·3

1-1·

62)

<0·0

011·

08 (1

·05-

1·12

)<0

·001

1·

08 (1

·05-

1·12

)<0

·001

Car

bohy

drat

es, t

otal

1·05

(1·0

2-1·

08)

0·00

41·

05 (1

·01-

1·08

)0·

006

0·94

(0·8

8-1·

01)

0·09

0·

99 (0

·98-

1·00

)0.

2

Mon

o- a

nd d

isac

char

ides

1·11

(1·0

5-1·

17)

<0·0

011·

10 (1

·04-

1·16

)0·

001

1·03

(0·9

6-1·

11)

0·4

1·

01 (0

·99-

1·02

)0.

6

Poly

sacc

hari

des

1·03

(0·9

8-1·

09)

0·3

1·03

(0·9

8-1·

09)

0·2

0·86

(0·7

9-0·

94)

0·00

1

0·97

(0·9

5-0·

99)

0·01

Fibr

es0·

81 (0

·49-

1·34

)0·

40·

85 (0

·51-

1·41

)0·

50·

25 (0

·13-

0·49

)<0

·001

0·

78 (0

·68-

0·89

)<0

·001

Alc

ohol

0·96

(0·8

9-1·

04)

0·4

0·95

(0·8

7-1·

03)

0·2

0·91

(0·8

4-0·

99)

0·03

0·

98 (0

·96-

1·00

)0·

03V

itam

ins

and

min

eral

s

C

alci

um1·

03 (1

·00-

1·05

)0·

031·

03 (1

·00-

1·05

)0·

030·

99 (0

·97-

1·02

)0·

7

1·00

(0·9

9-1·

01)

0.7

V

itam

in B

2 1·

24 (1

·07-

1·43

)0·

004

1·25

(1·0

8-1·

45)

0·00

31·

07 (0

·88-

1·29

)0·

5

1·03

(0·6

8-1·

56)

0.9

V

itam

in C

0·96

(0·8

1-1·

13)

0·6

0·98

(0·8

3-1·

16)

0·8

0·83

(0·6

9-1·

00)

0·05

0·

97 (0

·93-

1·00

)0.

08

Vit

amin

E

1·02

(1·0

0-1·

03)

0·05

1·02

(1·0

0-1·

03)

0·04

0·98

(0·9

6-1·

01)

0·2

0·

95 (0

·90-

1·00

)0.

05

Lyco

pene

1·00

(1·0

0-1·

00)

0·9

1·00

(1·0

0-1·

00)

0·97

1·00

(1·0

0-1·

00)

0·4

0·

97 (0

·92-

1·02

)0.

2

Flav

onoi

ds0·

51 (0

·27-

0·95

)0·

030·

53 (0

·28-

1·00

)0·

050·

36 (0

·18-

0·70

)0·

002

0·

98 (0

·97-

0·99

)0.

004

Glu

tam

ate

1·04

(1·0

1-1·

07)

0·02

1·04

(1·0

1-1·

07)

0·01

1·01

(0·9

7-1·

04)

0·6

1·

00 (1

·00-

1·01

)0.

3P

hyto

estr

ogen

s0·

99 (0

·93-

1·05

)0·

70·

98 (0

·93-

1·04

)0·

60·

98 (0

·93-

1·04

)0·

5

0·95

(0·8

4-1·

07)

0.4

a Adj

uste

d fo

r ag

e (a

t ons

et in

pat

ient

s; a

t que

stio

nnai

re in

con

trol

s), g

ende

r, a

nd e

duca

tion

b Bol

d in

dica

tes

Bon

ferr

oni s

igni

fica

nt v

alue

s of

p; B

onfe

rron

i adj

uste

d ɑ:

0·0

5/25

= 0

·002

c Adj

uste

d fo

r ag

e (a

t ons

et in

pat

ient

s; a

t que

stio

nnai

re in

con

trol

s), g

ende

r, e

duca

tion

, BM

I, cu

rren

t sm

okin

g, a

nd li

feti

me

phys

ical

act

ivit

yd A

djus

ted

for

age

(at o

nset

in p

atie

nts;

at q

uest

ionn

aire

in c

ontr

ols)

, gen

der,

edu

cati

on, B

MI,

curr

ent s

mok

ing,

life

tim

e ph

ysic

al a

ctiv

ity,

and

tota

l ene

rgy

inta

keA

LS

= a

myo

trop

hic

late

ral s

cler

osis

; CI =

con

fide

nce

inte

rval

; OR

= O

dds

rati

o


57


of saturated fat is independently associated with an increased risk of ALS (p=0·04), while a higher intake of alcohol is independently associated with a decreased risk of ALS (p=0·03). Also, a higher premorbid BMI is associated in this multivariate analysis with a decreased risk of ALS (p=0·01). Total energy intake is not significantly associated with risk of ALS in this model (p=0·09).No significant associations between nutrient intake and survival were found with multivariate Cox regression analysis (not shown). Several significant associations between nutrients and age at onset were identified. An interaction term of case-control status and the nutrient introduced into the model was, however, not significant for any of these associations; furthermore, the same associations were found when Cox regression was performed in controls using questionnaire completion as the event. These both findings indicate that the associations between nutrients and age at onset are an age-related effect and thus not disease-specific.

DISCUSSION

In the present population-based case-control study, we found an increased risk of sporadicALS with higher premorbid intake of total fat, saturated fat, trans fatty acids, and cholesterol and a low intake of alcohol. Furthermore, the pre-symptomatic daily energy intake in patients was significantly higher compared with controls, while pre-symptomatic BMI was significantly lower in patients. The combination of a positive association of a high total energy intake, a low premorbid BMI, a high fat intake and a low intake of alcohol, corrected for lifetime physical activity, supports a role for an altered energy metabolism prior to clinical onset of ALS.

Table 4.3 Adjusted ORs for the relationship between ALS and nutrient intakes in a multivariate model

Nutrient Adjusted ORa (95% CI) P value

Protein, vegetable 0·997 (0·991-1·004) 0·4

Fat, saturated 1·002 (1·000-1·004) 0·04

Polysaccharides 0·999 (0·998-1·000) 0·1

Fibres 0·998 (0·985-1·010) 0·7

Alcohol 0·999 (0·998-1·000) 0·03

Flavonoids 0·996 (0·987-1·004) 0·3

Total energy intake 1·000 (1·000-1·001) 0·09

Premorbid BMI 0·967 (0·944-0·992) 0·01a Adjusted for age (at onset in patients; at questionnaire in controls), gender, education, BMI, current smoking, lifetime physical activity, and total energy intake. ALS = amyotrophic lateral sclerosis; CI = confidence interval; OR = Odds ratio


CHAPTER 4

58

The finding that a higher intake of fat is associated with an increased risk of developing ALS corroborates observations in a population-based case-control study in ALS in Western Washington State.3 Another case-control study, however, found a contradicting result: a decreased risk of ALS with a higher intake of fat.7 Differences in study design may explain this discrepancy. The inclusion of only clinic-based patients may have caused referral bias in the latter study.33 Furthermore, in the Western Washington State study and the present study, only incident cases were included, and it is well-known that patient characteristics show large differences between an incident and prevalent cohort of ALS patients.1 This illustrates the importance of adopting a population-based approach when using a case-control design in ALS.Multiple studies have shown that ALS patients, after symptom onset, have an altered energy metabolism.34-38 In one study, the mean weight-adjusted resting energy expenditure was 15·7% higher than in a group of controls, and in another study all 11 familial ALS patients and 17 of the 33 sporadic ALS patients were hypermetabolic.34, 35 There is also a growing body of evidence that the mutant SOD1 mouse model of ALS shows metabolic alterations. Resting and total energy expenditure of G86R and G93A mice, when compared to wild-type littermates, were shown to be markedly increased, also in pre-symptomatic mice.39 In addition, increased lipolysis has recently been shown to occur in SOD1 mutant mice, also in pre-symptomatic mice.40 It has been suggested that mitochondrial uncoupling protein 3 (UCP3) plays a role in this increased energy expenditure, since higher levels of expression of UCP3 have been found both in an animal model of ALS and in human biopsies, while transgenic mice that overexpress UCP3 in muscles are lean and hyperphagic due to hypermetabolism through mitochondrial uncoupling.41-43 Our finding that pre-symptomatic daily energy intake in patients is higher, while pre-symptomatic BMI is lower, something which has also been shown previously in large cohort and case-control studies,44, 45 supports an increased energy expenditure in pre-symptomatic ALS patients. Since fat has a high caloric density, the higher premorbid intake of fat in ALS patients in the present study may be a compensatory mechanism for this increased energy expenditure to prevent weight and muscle loss. This may also explain the positive effect of hypercaloric enteral nutrition on survival in ALS patients in a recent phase 2 trial.46 A previous study, however, has shown that a high-fat diet itself increases resting energy expenditure, which may support a hypothesis that high intake of fat in pre-symptomatic ALS patients is not a compensation for increased energy expenditure, but may have, partly, caused the increased energy expenditure.47 It remains uncertain whether these findings are part of a disease-causing chain of events in ALS or whether they represent secondary phenomena. Our observations further emphasize the importance of a comparison in a future phase 3 trial: to establish whether a high-carbohydrate, high-caloric diet is to be preferred to a high-fat, high-caloric diet in ALS.46 Nevertheless, the present study lends support to the hypothesis that altered energy metabolism may already be present in pre-symptomatic ALS patients. There are several possible explanations for the observed decreased risk of ALS associated with a higher intake of alcohol. One previous population-based study could not identify


59


an association between alcohol consumption and ALS, but only 161 patients were included.48 Other, relatively small, studies have shown conflicting results but suffered from bias, since only clinic-based referral patients were included or there was no detailed record on alcohol consumption.49, 50 A previous study has shown that a lyophilized extract of red wine, which contains several antioxidant compounds, was able to block glutamate-induced apoptosis in cerebellar granule neurones.51 Furthermore, an in vivo experiment carried out on mutant SOD1 mice showed that survival in mice fed with lyophilized red wine was significantly increased compared to control, untreated animals. In our study, however, the association between intake of alcohol and the risk of ALS was independent of the intake of red wine (not shown), and so the association cannot be attributed only to the possible protective effect of antioxidants in red wine. Two previous case-control studies have shown that a high intake of polyunsaturated fatty acids (PUFA) is associated with a decreased risk of ALS.4, 7 PUFA have neuroprotective properties, since they exert beneficial effects on excitotoxicity, inflammation and oxidative stress.52 Also in Parkinson disease, a decreased risk has been found with a higher intake of PUFA.53 In our present study, we did not observe a significant association between intake of PUFA and risk of ALS; nor did we find an association between the risk of ALS and the intake of omega 3 fatty acids, which is a subtype of PUFA. In a recent cohort study, only a higher intake of this subtype was associated with a decreased risk of ALS.14 There is, however, also in our study, a trend towards a decreased risk of ALS with higher intake of PUFA (p trend = 0·1). Despite the relatively large study sample, the power may still have been too small to identify a significant association. In addition, the FFQs differ between the studies, which may have contributed to inconsistent results. The FFQ used in the present study covered all relevant sources of omega 3 fatty acids and other PUFA, including several types of fish, oils and supplements.The present study does not lend support to the hypothesis that dietary antioxidants have an independent protective effect on developing ALS, which has been suggested, since prior research showed a role for oxidative stress in the pathogenesis of ALS.3, 4, 6, 11, 13 Although in the univariate analysis, adjusted for confounders, a higher intake of flavonoids was associated with a decreased risk of ALS (p=0·002), in the multivariate analysis including other nutrients, this association was not significant (p=0·3). The present study, therefore, does not confirm previous findings that intake of vitamin E and the antioxidative carotenoids are inversely related to the risk of ALS.4, 11, 13 Our previous study which showed a possible protective effect of a high vitamin E intake in developing ALS only included patients ascertained in tertiary care centres.4 It has been demonstrated that ALS patients attending these referral centres do not represent a random sample of all ALS patients.33 A difference in vitamin E intake by these patients compared with non-referred patients may have led to biased results. The second study that showed a possible protective effect of vitamin E was a pooled analysis of five prospective cohort studies.11 In that study, however, there was a discrepancy between a lower ALS risk associated with higher dietary vitamin E intake and a lack of association with overall supplemental vitamin E intake. In our study, vitamin E intake was calculated from both dietary and supplemental intake as


CHAPTER 4

60

derived from the food-frequency questionnaire. We, therefore, cannot lend further support to a protective effect of vitamin E intake on developing ALS. Besides the strengths of our study, which include a relatively large study sample, the use of a validated questionnaire, a population-based setting, a control population representative of the general population, a correction for multiple comparisons, and a correction for many possible confounders, including physical activity, we also have to acknowledge limitations. Since ALS patients search actively for an explanation for their disease or may have an assumption about the underlying cause, case-control studies in ALS using questionnaires are inevitably prone to recall bias. Blinding the participants to the study hypotheses with an elaborate food frequency questionnaire, covering 199 food items, and the short time between date of diagnosis and date on which the questionnaire was completed (median 2·3 months) may have reduced this source of bias in our study. Another limitation is that bulbar symptoms may have affected usual dietary habits and, subsequently, how patients filled in the questionnaire, despite the fact that patients were asked to recall their dietary habits during the period before the onset of bulbar signs. A sensitivity analysis excluding bulbar onset patients did not essentially change results, suggesting that the identified associated dietary pattern was not the result of disease-related dietary changes.In conclusion, the combination of a positive association of a low premorbid BMI and a high fat intake, together with prior evidence from ALS SOD1 mouse models and earlier reports on premorbid BMI, supports a role for altered energy metabolism prior to clinical onset of ALS.


61


PANEL: RESEARCH IN CONTEXT

Evidence before this studyWe searched PubMed up to January 10, 2015, for reports in English, and excluding animal studies, with the terms "amyotrophic lateral sclerosis" and “consumption” or “diet” or “intake”, yielding 152 reports. We screened these reports for content and identified 15 papers addressing the relation between intake of food items or nutrients and the risk of ALS in a case-control or cohort design. 6 papers reported on a non-population-based case-control study, which is prone to referral bias.4, 5, 8, 10, 12, 54 Between 77 and 377 ALS patients were included in these studies, and except for a decreased risk of ALS with a higher intake of fruit8, 54 and vegetables8, 10 that was found in two studies, none of the associations was replicated in one of the other studies. Two papers reported on the same population-based case-control study with 161 included ALS patients, and showed that a higher fat and glutamate intake was associated with an increased risk of ALS.3, 48 7 papers reported on the results of the Cancer Prevention Study II cohort, and 5 of these papers pooled these data with data of 4 other cohorts, with a maximum of 1,153 documented ALS cases.6, 9, 11, 13, 14, 55, 56 Since the results are reported in separate papers, the analyses are not adjusted for multiple comparisons. These papers showed that a higher intake of chicken, ω-3 polyunsaturated fatty acid, cartenoids, and vitamin E are associated with a decreased risk of ALS.

Added value of this studyTo our knowledge, this is the largest population-based case-control study to date on the relation between premorbid dietary intake and the risk of sporadic ALS. We show that presymptomatic intake of fat and total daily energy intake in patients are higher, while presymptomatic BMI is significantly lower. This supports a role for altered energy metabolism prior to clinical onset of ALS, and provide further insight into the pathophysiological process of ALS. Further, this is the first study to show that alcohol is associated with a decreased risk of ALS.

Implications of all the available evidenceThe finding that a higher intake of fat is associated with an increased risk of developing ALS corroborates the observations in a prior population-based case-control study, and warrants further research on energy metabolism prior to clinical onset of ALS.3 The observation of a decreased risk of ALS with higher alcohol intake first needs replication by another study.


CHAPTER 4

62

REFERENCES


2. Kiernan MC, et al. Amyotrophic lateral sclerosis. Lancet. 2011; 377: 942-55.3. Nelson LM, et al. Population-based case-control study of amyotrophic lateral sclerosis in

western Washington State. II. Diet. Am J Epidemiol. 2000; 151: 164-73.4. Veldink JH, et al. Intake of polyunsaturated fatty acids and vitamin E reduces the risk of

developing amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry. 2007; 78: 367-71.5. Longnecker MP, et al. Dietary intake of calcium, magnesium and antioxidants in relation to

risk of amyotrophic lateral sclerosis. Neuroepidemiology. 2000; 19: 210-6.6. Ascherio A, et al. Vitamin E intake and risk of amyotrophic lateral sclerosis. Ann Neurol. 2005;

57: 104-10.7. Okamoto K, et al. Nutritional status and risk of amyotrophic lateral sclerosis in Japan.

Amyotroph Lateral Scler. 2007; 8: 300-4.8. Okamoto K, et al. Fruit and vegetable intake and risk of amyotrophic lateral sclerosis in Japan.

Neuroepidemiology. 2009; 32: 251-6.9. Morozova N, et al. Diet and amyotrophic lateral sclerosis. Epidemiology. 2008; 19: 324-37.10. Binazzi A, et al. An exploratory case-control study on spinal and bulbar forms of amyotrophic

lateral sclerosis in the province of Rome. Amyotroph Lateral Scler. 2009; 10: 361-9.11. Wang H, et al. Vitamin E intake and risk of amyotrophic lateral sclerosis: a pooled analysis of

data from 5 prospective cohort studies. Am J Epidemiol. 2011; 173: 595-602.12. Beghi E, et al. Coffee and amyotrophic lateral sclerosis: a possible preventive role. Am J

Epidemiol. 2011; 174: 1002-8.13. Fitzgerald KC, et al. Intakes of vitamin C and carotenoids and risk of amyotrophic lateral

sclerosis: pooled results from 5 cohort studies. Ann Neurol. 2013; 73: 236-45.14. Fitzgerald KC, et al. Dietary omega-3 polyunsaturated fatty acid intake and risk for

amyotrophic lateral sclerosis. JAMA Neurol. 2014; 71: 1102-10.15. Brooks BR, et al. El Escorial revisited: revised criteria for the diagnosis of amyotrophic lateral

sclerosis. Amyotroph Lateral Scler Other Motor Neuron Disord. 2000; 1: 293-9.16. DeJesus-Hernandez M, et al. Expanded GGGGCC hexanucleotide repeat in noncoding region

of C9orf72 causes chromosome 9p-linked FTD and ALS. Neuron. 2011; 72: 245-56.17. Netherlands Nutrition Centre. Zo eet Nederland 1992: resultaten van de

voedselconsumptiepeiling 1992 (in Dutch). (Dutch National Food Consumption Survey 1992). the Hague1993.

18. Netherlands Nutrition Centre. Zo eet Nederland: resultaten van de Voedselconsumptiepeiling 1997-1998 (in Dutch). (Dutch National Food Consumption Survey 1997-1998). The Hague1998.

19. Feunekes GI, et al. Relative and biomarker-based validity of a food-frequency questionnaire estimating intake of fats and cholesterol. Am J Clin Nutr. 1993; 58: 489-96.

20. van der Heijden L. Maten, gewichten en codenummers 2003. Informatie voor Voeding en Dietiek BohnStafleuvanLoghum,2013.

21. Stichting Nederlands Voedingsstoffenbestand. NEVO-tabel: Nederlands Voedingsstoffenbestand 2006 (in Dutch). (Dutch Food Composition Database 2006). The Hague2006.

22. HertogMG,etal.Intakeofpotentiallyanticarcinogenicflavonoidsandtheirdeterminantsinadults in The Netherlands. Nutr Cancer. 1993; 20: 21-9.

23. HertogMG, et al. Dietary antioxidant flavonoids and risk of coronary heart disease: theZutphen Elderly Study. Lancet. 1993; 342: 1007-11.

24. USDA-Iowa State University Database. 2012. Online source.25. Rhodes J, et al. A survey of the monosodium glutamate content of foods and an estimation of

the dietary intake of monosodium glutamate. Food Addit Contam. 1991; 8: 663-72.26. Loliger J. Function and importance of glutamate for savory foods. J Nutr. 2000; 130: 915S-20S.


63


27. Skurray GRP, N.;. 1-glutamic acid content of fresh and processed foods. Food Chemistry. 1988; 27: 4.

28. Yamaguchi S and Ninomiya K. Umami and food palatability. J Nutr. 2000; 130: 921S-6S.29. Glutamate content of foods. 2012. Online source.30. Black AE. Critical evaluation of energy intake using the Goldberg cut-off for energy

intake:basal metabolic rate. A practical guide to its calculation, use and limitations. Int J Obes Relat Metab Disord. 2000; 24: 1119-30.

31. Huisman MH, et al. Lifetime physical activity and the risk of amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry. 2013; 84: 976-81.

32. Willett WC, et al. Adjustment for total energy intake in epidemiologic studies. Am J Clin Nutr. 1997; 65: 1220S-8S; discussion 9S-31S.

33. Sorenson EJ, et al. Effect of referral bias on assessing survival in ALS. Neurology. 2007; 68: 600-2.34. Funalot B, et al. High metabolic level in patients with familial amyotrophic lateral sclerosis.

Amyotroph Lateral Scler. 2009; 10: 113-7.35. Desport JC, et al. Factors correlated with hypermetabolism in patients with amyotrophic

lateral sclerosis. Am J Clin Nutr. 2001; 74: 328-34.36. Kasarskis EJ, et al. Nutritional status of patients with amyotrophic lateral sclerosis: relation to

the proximity of death. Am J Clin Nutr. 1996; 63: 130-7.37. Dupuis L, et al. Energy metabolism in amyotrophic lateral sclerosis. Lancet Neurol. 2011; 10:

75-82.38. Genton L, et al. Nutritional state, energy intakes and energy expenditure of amyotrophic

lateral sclerosis (ALS) patients. Clin Nutr. 2011; 30: 553-9.39. Dupuis L, et al. Evidence for defective energy homeostasis in amyotrophic lateral sclerosis:

benefitofahigh-energydietinatransgenicmousemodel.Proc Natl Acad Sci U S A. 2004; 101: 11159-64.

40. Dodge JC, et al. Metabolic signatures of amyotrophic lateral sclerosis reveal insights into disease pathogenesis. Proc Natl Acad Sci U S A. 2013; 110: 10812-7.

41. Li J, et al. Reducing systemic hypermetabolism by inducing hypothyroidism does not prolong survival in the SOD1-G93A mouse. Amyotroph Lateral Scler. 2012; 13: 372-7.

42. Dupuis L, et al. Up-regulation of mitochondrial uncoupling protein 3 reveals an early muscular metabolic defect in amyotrophic lateral sclerosis. FASEB J. 2003; 17: 2091-3.

43. Clapham JC, et al. Mice overexpressing human uncoupling protein-3 in skeletal muscle are hyperphagic and lean. Nature. 2000; 406: 415-8.

44. O'Reilly EJ, et al. Premorbid body mass index and risk of amyotrophic lateral sclerosis. Amyotroph Lateral Scler Frontotemporal Degener. 2013; 14: 205-11.

45. Scarmeas N, et al. Premorbid weight, body mass, and varsity athletics in ALS. Neurology. 2002; 59: 773-5.

46. Wills AM, et al. Hypercaloric enteral nutrition in patients with amyotrophic lateral sclerosis: a randomised, double-blind, placebo-controlled phase 2 trial. Lancet. 2014; 383: 2065-72.

47. Ebbeling CB, et al. Effects of dietary composition on energy expenditure during weight-loss maintenance. JAMA. 2012; 307: 2627-34.

48. Nelson LM, et al. Population-based case-control study of amyotrophic lateral sclerosis in western Washington State. I. Cigarette smoking and alcohol consumption. Am J Epidemiol. 2000; 151: 156-63.

49. Okamoto K, et al. Lifestyle factors and risk of amyotrophic lateral sclerosis: a case-control study in Japan. Ann Epidemiol. 2009; 19: 359-64.

50. Veldink JH, et al. Physical activity and the association with sporadic ALS. Neurology. 2005; 64: 241-5.

51. Esposito E, et al. Lyophilized red wine administration prolongs survival in an animal model of amyotrophic lateral sclerosis. Ann Neurol. 2000; 48: 686-7.

52. Zhang W, et al. Omega-3 polyunsaturated fatty acids in the brain: metabolism and neuroprotection. Front Biosci (Landmark Ed). 2011; 16: 2653-70.

53. de Lau LM, et al. Dietary fatty acids and the risk of Parkinson disease: the Rotterdam study. Neurology. 2005; 64: 2040-5.


CHAPTER 4

64

54. Jin Y, et al. Dietary intake of fruits and beta-carotene is negatively associated with amyotrophic lateral sclerosis risk in Koreans: a case-control study. Nutr Neurosci. 2014; 17: 104-8.

55. FondellE,etal.Dietaryfiberandamyotrophic lateralsclerosis:resultsfrom5largecohortstudies. Am J Epidemiol. 2014; 179: 1442-9.

56. FondellE,etal.Magnesiumintakeandriskofamyotrophiclateralsclerosis:resultsfromfivelarge cohort studies. Amyotroph Lateral Scler Frontotemporal Degener. 2013; 14: 356-61.


65



Figure S4.1 Odds ratios for the relationship between ALS and quintiles of nutrient intake. Adjusted for energy intake, age (at onset in patients; at questionnaire in controls), gender, BMI, education, current smoking, and lifetime physical activity.



Meinie Seelen1*, Mark H.B. Huisman1*, Loes van Boxmeer1*, Anne E. Visser1, Perry T.C. van Doormaal1, Joost Raaphorst2, Anneke J. van der Kooi3,

Marianne de Visser3, Roel C.H. Vermeulen4, Leonard H. van den Berg1, Jan H. Veldink1

1 Department of Neurology, Brain Center Rudolf Magnus, University Medical Center Utrecht, the Netherlands.

2 Department of Neurology, Donders Institute for Brain, Cognition and Behaviour, Center for Neuroscience, Radboud University Nijmegen Medical Center, the Netherlands.

3 Department of Neurology, Amsterdam Medical Center, University of Amsterdam, the Netherlands.

4 Environmental Epidemiology Division, Institute for Risk Assessment Sciences, Utrecht University, the Netherlands.

* These authors contributed equally.

In preparation

Occupational exposure to diesel motor exhaust

increases the risk of amyotrophic lateral sclerosis

CHAPTER 5

PART II - ENVIRONMENT


CHAPTER 5

68

SUMMARY

Background The association of amyotrophic lateral sclerosis (ALS) with occupations has been studied extensively, since occupations may serve as a surrogate for a variety of environmental exposures, possibly leading to the development of ALS. However, limited data is available concerning direct and objective investigation of occupational exposure to environmental toxins. In this study, occupational exposures and their association with ALS risk were assessed through the application of job exposure matrices (JEMs), a valid and objective exposure assessment tool.

MethodsA large population-based, case-control study was conducted in two independent populations in Europe: 662 patients and 2,152 controls in the Netherlands, and 142 patients and 255 controls in the Republic of Ireland. Lifetime occupational history was obtained using a structured questionnaire, and coded according to the International Standard Classification of Occupations (ISCO). Job exposure matrices, assigning no, low, or high exposure for 17 different agents, were applied to determine cumulative levels of exposure before onset of disease. Odds ratios (OR) of ALS risk were estimated by multivariate logistic regression, adjusted for potential confounders. Finally, a meta-analysis of both populations was performed.

FindingsCumulative occupational exposure to diesel motor exhaust was associated with an increased risk of ALS (OR 1·10, 95% CI 1·03-1·18, p=0·004) in both populations. No association of ALS risk was found for the groups of mineral dust, organic dust, pesticides, metals or solvents.

InterpretationIn this study we showed, using a rigorous and population-based design, that occupational exposure to diesel motor exhaust is a risk factor for developing ALS.


69

Occupational exposures and ALS risk

INTRODUCTION

Sporadic amyotrophic lateral sclerosis (ALS) is considered to be a complex disease, in which both genetic and environmental factors determine disease susceptibility and outcome.1 In order to identify causative environmental factors, the association of ALS with occupations has been studied extensively,2-13 since occupations may serve as a surrogate for a variety of exogenous exposures (i.e. pesticides, metals, solvents, gasses and fumes). Unfortunately, these studies faced several challenges. Numbers of cases and controls per occupation were often too low to detect associations, while on the other hand most of the associations that were identified could not be replicated. Directly investigating (past) exposure to selected environmental agents instead of occupations was often limited by the exposure assessment method.14, 15 Examples are self-reported exposures that readily lead to differential responder bias, and the use of a job history registry, which is often inaccurate and incomplete.A job exposure matrix (JEM) is recognized as an objective, valid and agent-specific method for exposure assessment in case-control studies.16-18 A JEM enables linking of occupations to profiles of environmental exposures by providing (semi-)quantitative assessments of exogenous exposures for each occupation. The application of a JEM in ALS has only been applied in exposure studies of electric shocks and magnetic fields.19-21 Patients and controls are asked to fill in all the occupations they have held during life, without any clue as to what hypotheses will be tested, which largely avoids recall bias.22

The aim of this large, population-based case-control study, performed in two independent populations, was to determine the association between lifetime occupational exposure to a wide range of agents and the risk of ALS using a JEM as an unbiased, objective, and semi-quantitative exposure assessment method.

METHODS

The Netherlands: Prospective ALS study the Netherlands (PAN) A large population-based, case-control study was conducted in the Netherlands, between January 2006 and December 2010, entitled the “Prospective ALS study the Netherlands” (PAN).23 All newly diagnosed patients, with possible, probable (laboratory-supported) or definite ALS according to the revised El Escorial Criteria, were selected.24 Medical records of all patients were scrutinized to confirm the appropriateness of the diagnosis and to exclude ALS mimic syndromes or other clinical conditions. Every patient who had a first, second or third degree family member with ALS was defined as having familial ALS, and was excluded. Complete case ascertainment was ensured by continuous recruitment through multiple sources: 1) Neurologists, most ALS patients visit one of the tertiary referral centers of the ALS center the Netherlands on at least one occasion; 2) Consultants in rehabilitation medicine; 3) the Dutch Neuromuscular Patient Association; and 4) ALS website.


CHAPTER 5

70

Population-based controls were selected from the register of the general practitioner (GP) taking care of the patient with ALS. The GP was asked to select individuals, matched to the patient for gender and age (plus or minus five years), from his register in alphabetical order, starting at the surname of the ALS patient. Spouses or blood-relatives of the patient were excluded to prevent overmatching. The institutional review board of the University Medical Center Utrecht Ethics Committee approved this study. Informed consent was obtained from all participants.

Data ascertainmentParticipants were asked to fill in a structured questionnaire on their lifetime occupational history, including military service and periods spent as a homemaker. For each occupation the number of years and the hours per week employed in that job were recorded. If the questionnaire was not entirely completed or if data were found to be inconsistent, participants were approached by telephone to complete or correct the data. Information about education, body mass index (BMI), cigarette smoking and alcohol use was also obtained from this questionnaire. To ensure blinding, all questionnaires were coded prior to processing and analysis. Survival status of patients was recorded through the civil registry, the general practitioner and the motor neuron disease association. Among patients, only data before symptom onset were analysed.

Ireland: Irish ALS register A second independent population-based case-control study was performed in the Republic of Ireland between May 2011 and June 2014, through the Irish ALS register. Details of the Irish ALS Register have been published previously.25 Briefly, the Irish ALS Register was used to identify Irish residents diagnosed with suspected, possible, probable or definite ALS according to the El Escorial criteria.24 Most patients attended the Beaumont Hospital motor neuron disease clinic in Dublin. A minority of patients was seen in other neurology clinics or was contacted to participate through the Irish Motor Neurone Disease Association (IMNDA). For this study, we excluded patients with ALS mimic syndromes and familial ALS, which was defined as patients with a first, second or third degree family member with ALS.Population-based controls were selected in the same way as in the Netherlands, by the GP of the ALS patient and were individually matched for gender, age (plus or minus five years) and location of current residence. Spouses and blood-relatives of ALS patients were excluded to prevent overmatching.The Irish ALS Register complies with Irish Data protection legislation (1988 and 2003), and has been approved by Beaumont Hospital Ethics Committee (02/28 and 05/49). Verbal and written consent is obtained from all participants for inclusion on the Irish ALS Register.


71


Data ascertainmentIn Ireland, all patients were visited for a personal interview using the same structured questionnaire on lifetime occupational history as described above in the Netherlands. The questionnaires were handled equally: they were coded prior to processing and analysis, and only the data before onset of symptoms was used for patients. Survival status of patients was recorded through the civil registry, the general practitioner and the motor neuron disease association.

Classification of occupationsAll occupations were coded according to the International Standard Classification of Occupations (ISCO) adopted by the International Labor Organization (ILO), a United Nations specialized agency.26 The ISCO provides a systematic classification structure covering the occupations of the whole civilian working population. Both the 1968 version as the 1988 versions of the ISCO were used. The classification structure of the ISCO-68 has four levels, providing successively finer detail, as follows: major groups (8), minor groups (83), unit groups (284) and occupational categories (1,506). Since the ISCO-68 lacks code numbers for military services, armed forces and homemakers, we added supplemental major categories for these occupations. The ISCO-88 consists of ten major groups, subdivided into sub-major groups (28), minor groups (116), and unit groups (390).

Exposure assessmentExposures were estimated by using two general population job-exposure matrices: DOM-JEM, and ALOHA-JEM. The DOM-JEM40,41 is based on five-digit ISCO-68 codes (occupational categories), and the ALOHA-JEM42 on four-digit ISCO-88 codes (unit groups). The JEM’s were created by occupational exposure experts, and assign exposure intensity scores of no exposure, low or high exposure levels to each ISCO code. Cumulative exposure is calculated by summing the product of the intensity and duration (years) for all reported job periods over the entire working career. The exposure intensity scores of none, low and high were transformed to 0, 1 and 2 to achieve a more balanced weighting between intensity and duration in the calculation of cumulative exposure.17 exposures were assessed through the DOM-JEM and ALOHA-JEM in six main groups: mineral dust (silica, asbestos), organic dust (animal contact, endotoxin), pesticides (herbicides, insecticides), gasses and fumes (polycyclic aromatic hydrocarbon (PAH), diesel motor exhaust (DME)), metals (chromium, nickel) and solvents (aromatic solvents, chlorinated solvents).

Statistical analysisDifferences in baseline characteristics between patients and controls were determined using χ2 test for categorical variables and Mann-Withney U test for continuous variables. For each participant, and each exposure from the Job Exposure Matrices, a lifetime exposure index was calculated by the following formula:


CHAPTER 5

72

where k represents a job from the lifetime occupational history. For each exposure, participants were classified as never exposed (0), low lifetime exposure index (1), or high lifetime exposure index (2). Low and high exposure represent respectively a lower or higher exposure than the median of the cumulative exposure score among the exposed controls. Multivariate logistic regression analysis was used to estimate odds ratios (ORs) for the association with ALS for low and high exposure compared with no exposure, adjusting for covariates age, gender, education, BMI, smoking and alcohol use. Age was defined as age at onset in patients and age at date on which the questionnaire was completed in controls. Initially, logistic regression analyses were performed separately for the Dutch and Irish population data. p values for trend (dose-response) were obtained using a log-transformed lifetime exposure index as a continuous variable. Secondly, fixed effects meta-analysis was performed comparing the population data by using a logistic regression model including the log-transformed lifetime exposure index. To determine whether especially recent exposure may act as a trigger in developing ALS, additional multivariate logistic regression analyses of the last job were performed. Odds ratios for ALS associated with the exposure intensity score during the last job were determined for exposures that had a p value ≤ 0·2 in the meta-analysis of cumulative exposure. Finally, in a cox regression survival analysis the relation between the exposure index of each exposure and disease duration was investigated, with age, gender, site of onset, BMI and current smoking as covariates. The same method was used to evaluate the effect of exposure on the age at onset of ALS patients, adjusted for gender and site of onset. All tests were two-sided, and a Bonferroni correction was applied to the alpha level to adjust for multiple comparisons. Since the exposures within the main groups were highly correlated, adjustment was applied for the six main groups (Bonferroni adjusted p value: 0·05/6 = 0·008).

RESULTS

The NetherlandsIn the Netherlands, of the questionnaires send to 782 patients, 662 (85%) were returned. Gender, median age at onset, and site of onset did not differ significantly between responders and non-responders. 2,332 population-based controls were selected from the GP’s register, and 2,152 of these returned their questionnaire (response rate 92%). Table 5.1 shows the baseline characteristics of sporadic ALS patients and controls included in this study. Cases and controls were similar for the matching variables age and gender. Furthermore, ALS cases more often had a lower educational level, a lower BMI, smoked more often and consumed less alcohol compared to controls.


73


Table 5.2 shows the ORs for ALS risk associated with cumulative exposures divided in no, low and high exposure groups. Cumulative exposure to DME was the only exposure that showed a significant linear trend for ALS risk (p=0·03) in the Netherlands, which was no longer significant after correction for multiple testing.

IrelandIn Ireland, 164 patients with ALS and 271 controls were recruited and interviewed. Age and gender were similar for cases and controls. Comparable to the Dutch dataset, cases more often had a lower educational level, a lower BMI, smoked more often and consumed less alcohol compared to controls. Compared to the Dutch ALS cases, Irish ALS cases had a slightly later age at onset, less often a bulbar site of onset and more often a definite El Escorial classification.In Ireland, cumulative exposure to DME showed, a linear trend for an increased risk of ALS (p=0·02). Moreover, a significant linear trend between ALS risk and exposures was also observed for mineral dust (p=0·03), silica (p=0·04) and chromium (p=0·03).


The Netherlands Ireland

Patients Controls Patients Controls

Variable (n=662) (n=2152) (n=142) (n=255)

Male, n (%) 410 (62) 1244 (58) 86 (61) 148 (58)

Age, y, median (IQR)a 63·4 (57-70) 62·9 (57-70) 64·5 (57-71) 67·5 (59-73)

Age at diagnosis, y, median (IQR) 64·6 (58-71) 66·5 (58-73)

Bulbar onset, n (%) 213 (32) 30 (21)

El Escorial classification

Definite, n (%) 116 (18) 65 (46)

Probable, n (%) 407 (61) 35 (25)

Possible, n (%) 125 (19) 27 (19)

Missing, n (%) 14 (2) 14 (10)

Education, n (%)*

No education / Primary school 61 (9) 134 (6) 48 (34) 68 (27)

Middle school / High school 438 (67) 1417 (66) 71 (50) 137 (54)

College / University 162 (24) 599 (28) 22 (16) 50 (19)

Body Mass Index, median (IQR)*† 24·2 (22-26) 25·6 (24-28) 25·3 (23-28) 26·3 (24-30)

Current smoker, n (%)* 134 (20) 287 (13) 21 (15) 25 (10)

Current alcohol, n (%)* 492 (74) 1832 (85) 96 (68) 195 (77)a Age at onset in patients, and age on which the questionnaire was completed in controls* Significant difference between patients and controls at a level of <0·05 for the Dutch population† Significant difference between patients and controls at a level of <0·05 for the Irish population


CHAPTER 5

74

Tab

le 5

.2 A

LS

risk

ass

ocia

ted

wit

h cu

mul

ativ

e ex

posu

res

for

the

Dut

ch a

nd Ir

ish

popu

lati

on

Th

e N

eth

erla

nds

Irel

and

Ex

pos

ure

Cu

mu

lati

ve

exp

osu

reC

ases

, n (%

)C

ontr

ols,

n

(%)

OR

(95%

CI)

†C

um

ula

tive

ex

pos

ure

Cas

es, n

(%)

Con

trol

s, n

(%

)O

R (9

5% C

I)†

Min

eral

dus

tN

ever

360

(54)

1210

(56)

1·00

(Ref

)N

ever

73 (5

2)15

4 (6

0)1·

00 (R

ef) *

≤17·

3013

9 (2

1)47

1 (2

2)0·

92 (0

·73-

1·17

)≤1

5·00

20 (1

4)51

(20)

0·81

(0·4

3-1·

53)

>17·

3016

2 (2

5)47

1 (2

2)1·

01 (0

·78-

1·29

)>1

5·00

48 (3

4)50

(20)

2·01

(1·0

9-3·

68)

Silic

aN

ever

606

(92)

2000

(93)

1·00

(Ref

)N

ever

113

(80)

227

(89)

1·00

(Ref

) *

≤17·

0029

(4·4

)76

(3·5

)1·

27 (0

·80-

2·01

)≤6

·25

8 (5

·7)

14 (5

·5)

0·99

(0·3

7-2·

64)

>17·

0026

(3·9

)76

(3·5

)0·

90 (0

·55-

1·45

)>6

·25

20 (1

4)14

(5·5

)2·

29 (1

·03-

5·09

)

Asb

esto

sN

ever

521

(79)

1745

(81)

1·00

(Ref

)N

ever

106

(75)

194

(76)

1·00

(Ref

)

≤18·

0075

(11)

203

(9·4

)1·

11 (0

·82-

1·51

)≤5

·00

9 (6

·4)

31 (1

2)0·

39 (0

·16-

0·93

)

>18·

0065

(9·8

)20

4 (9

·5)

0·98

(0·7

1-1·

36)

>5·0

026

(18)

30 (1

2)1·

39 (0

·73-

2·64

)

Org

anic

dus

tN

ever

440

(67)

1444

(67)

1·00

(Ref

)N

ever

79 (5

6)14

9 (5

8)1·

00 (R

ef)

≤12·

4493

(14)

354

(16)

0·85

(0·6

5-1·

11)

≤10·

1520

(14)

53 (2

1)0·

65 (0

·34-

1·24

)

>12·

4412

8 (1

9)35

4 (1

6)1·

10 (0

·86-

1·41

)>1

0·15

42 (3

0)53

(21)

1·36

(0·7

8-2·

39)

Ani

mal

con

tact

sN

ever

615

(93)

2007

(93)

1·00

(Ref

)N

ever

110

(78)

217

(85)

1·00

(Ref

)

≤16·

1825

(3·8

)73

(3·4

)1·

14 (0

·71-

1·85

)≤1

5·25

12 (8

·5)

19 (7

·5)

1·10

(0·4

8-2·

49)

>16·

1821

(3·2

)72

(3·3

)0·

89 (0

·53-

1·49

)>1

5·25

19 (1

4)19

(7·5

)1·

84 (0

·88-

3·87

)

End

otox

inN

ever

514

(78)

1697

(79)

1·00

(Ref

)N

ever

100

(71)

200

(78)

1·00

(Ref

)

≤8·5

562

(9·4

)22

8 (1

1)0·

86 (0

·62-

1·17

)≤1

6·00

21 (1

5)28

(11)

1·52

(0·7

9-2·

94)

>8·5

585

(13)

227

(11)

1·15

(0·8

7-1·

53)

>16·

0020

(14)

27 (1

1)1·

41 (0

·70-

2·84

)

† A

djus

ted

for

age,

gen

der,

edu

cati

on, b

ody

mas

s in

dex

curr

ent

smok

ing

and

alco

hol u

se· *

P v

alue

for

line

ar t

rend

<0·

05. N

o p

valu

es <

0·00

8. P

AH

, pol

ycyc

lic a

rom

atic

hy

droc

arbo

n; D

ME

, die

sel m

otor

exh

aust

. Cut

-off

val

ues

for

low

and

hig

h cu

mul

ativ

e ex

posu

re a

re b

ased

on

the

med

ian

valu

e of

the

cont

rols

.


75


Tab

le 5

.2 (c

onti

nued

)

Th

e N

eth

erla

nds

Irel

and

Ex

pos

ure

Cu

mu

lati

ve

exp

osu

reC

ases

, n

(%)

Con

trol

s, n

(%

)O

R (9

5% C

I)†

Cu

mu

lati

ve

exp

osu

reC

ases

, n

(%)

Con

trol

s, n

(%

)O

R (9

5% C

I)†

Pest

icid

esN

ever

588

(89)

1955

(91)

1·00

(Ref

)N

ever

111

(79)

221

(87)

1·00

(Ref

)

≤28·

0031

(4·7

)98

(4·6

)1·

09 (0

·70-

1·68

)≤2

1·00

18 (1

3)17

(6·7

)2·

29 (1

·09-

4·83

)

>28·

0042

(6·4

)99

(4·6

)1·

17 (0

·79-

1·73

)>2

1·00

12 (8

·5)

17 (6

·7)

1·06

(0·4

5-2·

49)

Her

bici

des

Nev

er61

3 (9

3)20

37 (9

5)1·

00 (R

ef)

Nev

er12

3 (8

7)23

4 (9

2)1·

00 (R

ef)

≤31·

5026

(3·9

)58

(2·7

)1·

51 (0

·92-

2·47

)≤1

0·00

10 (7

·1)

12 (4

·7)

1·39

(0·5

6-3·

48)

>31·

5022

(3·3

)57

(2·6

)0·

99 (0

·59-

1·67

)>1

0·00

8 (5

·7)

9 (3

·5)

1·11

(0·3

8-3·

25)

Inse

ctic

ides

Nev

er59

2 (9

0)19

69 (9

2)1·

00 (R

ef)

Nev

er11

1 (7

9)22

2 (8

7)1·

00 (R

ef)

≤28·

0029

(4·4

)93

(4·3

)1·

07 (0

·68-

1·67

)≤2

2·00

18 (1

3)17

(6·7

)2·

30 (1

·09-

4·85

)

>28·

0040

(6·1

)90

(4·2

)1·

24 (0

·83-

1·86

)>2

2·00

12 (8

·5)

16 (6

·3)

1·13

(0·4

8-2·

67)

Gas

ses

and

fum

esN

ever

262

(40)

837

(39)

1·00

(Ref

)N

ever

46 (3

3)87

(34)

1·00

(Ref

)

≤18·

0018

1 (2

7)65

6 (3

1)0·

88 (0

·70-

1·10

)≤2

6·38

42 (3

0)84

(33)

0·86

(0·4

8-1·

52)

>18·

0021

8 (3

3)65

9 (3

1)0·

92 (0

·73-

1·18

)>2

6·38

53 (3

8)84

(33)

0·96

(0·5

2-1·

78)

PAH

Nev

er58

8 (8

9)19

32 (9

0)1·

00 (R

ef)

Nev

er11

5 (8

2)20

6 (8

1)1·

00 (R

ef)

≤8

·26

43 (6

·5)

110

(5·1

)1·

18 (0

·81-

1·73

)≤4

·13

8 (5

·7)

25 (9

·8)

0·53

(0·2

2-1·

28)

>8·2

630

(4·5

)11

0 (5

·1)

0·75

(0·4

8-1·

16)

>4·1

318

(13)

24 (9

·4)

1·13

(0·5

6-2·

28)

DM

EN

ever

489

(74)

1686

(78)

1·00

(Ref

) *N

ever

86 (6

1)18

7 (7

3)1·

00 (R

ef) *

≤16·

3381

(12)

233

(11)

1·17

(0·8

7-1·

56)

≤17·

5025

(18)

34 (1

3)1·

86 (0

·98-

3·55

)

>16·

3391

(14)

233

(11)

1·28

(0·9

6-1·

71)

>17·

5030

(21)

34 (1

3)2·

04 (1

·06-

3·93

)

† A

djus

ted

for

age,

gen

der,

edu

cati

on, b

ody

mas

s in

dex

curr

ent s

mok

ing

and

alco

hol u

se. *

P v

alue

for

linea

r tr

end

<0·0

5. N

o p

valu

es <

0·00

8.PA

H, p

olyc

yclic

aro

mat

ic h

ydro

carb

on; D

ME

, die

sel m

otor

exh

aust

. Cut

-off

val

ues

for

low

and

hig

h cu

mul

ativ

e ex

posu

re a

re b

ased

on

the

med

ian

valu

e of

the

cont

rols

.


CHAPTER 5

76

Tab

le 5

.2 (c

onti

nued

)

Th

e N

eth

erla

nds

Irel

and

Ex

pos

ure

Cu

mu

lati

ve

exp

osu

reC

ases

, n

(%)

Con

trol

s, n

(%

)O

R (9

5% C

I)†

Cu

mu

lati

ve

exp

osu

reC

ases

, n

(%)

Con

trol

s, n

(%

)O

R (9

5% C

I)†

Met

als

Nev

er53

8 (8

1)18

00 (8

4)1·

00 (R

ef)

Nev

er10

7 (7

6)21

1 (8

3)1·

00 (R

ef)

≤33·

3063

(9·5

)17

6 (8

·2)

1·16

(0·8

3-1·

60)

≤7·5

12 (8

·5)

22 (8

·6)

1·08

(0·4

7-2·

45)

>33·

3060

(9·1

)17

6 (8

·2)

1·06

(0·7

5-1·

49)

>7·5

22 (1

6)22

(8·6

)1·

87 (0

·94-

3·72

)

Chr

omiu

m

Nev

er60

9 (9

2)20

07 (9

3)1·

00 (R

ef)

Nev

er12

9 (9

2)24

7 (9

7)1·

00 (R

ef) *

≤13·

0019

(2·9

)73

(3·4

)0·

82 (0

·48-

1·40

)≤9

·53

(2·1

)4

(1·6

)1·

92 (0

·36-

10·3

)

>13·

0033

(5·0

)72

(3·3

)1·

35 (0

·86-

2·12

)>9

·59

(6·4

)4

(1·6

)3·

36 (0

·97-

11·6

)

Nic

kel

Nev

er62

3 (9

4)20

26 (9

4)1·

00 (R

ef)

Nev

er13

4 (9

5)24

8 (9

7)1·

00 (R

ef)

≤13·

6313

(2·0

)63

(2·9

)0·

68 (0

·36-

1·26

)≤8

·00

2 (1

·4)

4 (1

·6)

1·18

(0·1

8-7·

72)

>13·

6325

(3·8

)63

(2·9

)1·

16 (0

·70-

1·91

)>8

·00

5 (3

·5)

3 (1

·2)

2·07

(0·4

7-9·

22)

Solv

ents

Aro

mat

ic s

olve

nts

Nev

er50

7 (7

7)16

17 (7

6)1·

00 (R

ef)

Nev

er90

(64)

189

(74)

1·00

(Ref

)

≤20·

0079

(12)

258

(12)

0·91

(0·6

8-1·

22)

≤14·

3820

(14)

33 (1

3)1·

19 (0

·60-

2·33

)

>20·

0075

(11)

254

(12)

0·79

(0·5

8-1·

07)

>14·

3831

(22)

33 (1

3)1·

83 (0

·99-

3·40

)

Chl

orin

ated

sol

vent

sN

ever

542

(82)

1743

(81)

1·00

(Ref

)N

ever

106

(75)

205

(80)

1·00

(Ref

)

≤26·

4059

(8·9

)20

5 (9

·5)

0·86

(0·6

2-1·

19)

≤9·8

816

(11)

25 (9

·8)

1·39

(0·6

8-2·

82)

>26·

4060

(9·1

)20

4 (9

·5)

0·84

(0·6

0-1·

17)

>9·8

819

(14)

25 (9

·8)

1·37

(0·6

9-2·

72)

† A

djus

ted

for

age,

gen

der,

edu

cati

on, b

ody

mas

s in

dex

curr

ent s

mok

ing

and

alco

hol u

se. *

P v

alue

for

linea

r tr

end

<0·0

5. N

o p

valu

es <

0·00

8.PA

H, p

olyc

yclic

aro

mat

ic h

ydro

carb

on; D

ME

, die

sel m

otor

exh

aust

. Cut

-off

val

ues

for

low

and

hig

h cu

mul

ativ

e ex

posu

re a

re b

ased

on

the

med

ian

valu

e of

the

cont

rols

.


77


However, these exposures were represented by small numbers, with less than ten cases or controls per exposure group for silica and chromium, and the effect was not consistent with the Dutch data. Furthermore, these linear trends were no longer significant after correction for multiple testing.

Meta-analysisIn the meta-analysis combining both populations, DME was the only exposure that was significantly associated with ALS risk (OR 1·10, 95% CI 1·03-1·18, p=0·004), depicted in Figure 5.1. This association remained significant after correction for multiple testing (p<0·008). Performing random effects instead of fixed effects meta-analysis of DME, similar results for the association with ALS risk were found (OR 1·12, 95% CI 1·01-1·24). None of the other cumulative exposures (i.e. within the main groups of mineral dust, organic dust, pesticides, metals, solvents) showed a significant altered risk of developing ALS.

Additional analysisIn the analysis of exposures during the last job performed, we assessed mineral dust, chromium, DME, pesticides and insecticides, which all had a p value ≤0·2 in the meta-analysis of cumulative exposures. In this last job analysis, none of the exposures reached a significant level <0·05, indicating that recent exposure may not be more important than overall exposure (Figure S5.1).Cox regression survival analyses showed that higher exposure to DME was associated with a shorter survival (meta-analysis: HR 1·09, 95% CI 1·02-1·16, p=0·01), however this association was no longer significant after correction for multiple testing. None of the other exposures modified disease duration. Analysis of age at onset in patients with ALS showed that DME exposure was associated with a later age at onset (62 versus 65 years, HR 0·93, 95% CI 0·88-0·98, p=0·005). To determine whether this effect was specific for patients, or also valid for age at questionnaire for controls, an additional analysis was performed: an interaction term of diagnosis and DME exposure was introduced into the model, with questionnaire completion as the event in controls (HR 0·95, p=0·11). This analysis indicates that the association between DME exposure and age at onset is an age related effect and not per se disease related. None of the other exposures showed an association with age at onset.


CHAPTER 5

78

Mineral dust Organic dust

Pesticides Gasses and fumes

Metals Solvents

Asbestos (Total)

Asbestos (IRE)

Asbestos (NL)

Silica (Total)

Silica (IRE)

Silica (NL)

Mineral dust (Total)

Mineral dust (IRE)

Mineral dust (NL)

Endotoxin (Total)

Endotoxin (IRE)

Endotoxin (NL)

Animal contacts (Total)

Animal contacts (IRE)

Animal contacts (NL)

Organic dust (Total)

Organic dust (IRE)

Organic dust (NL)

Insecticides (Total)

Insecticides (IRE)

Insecticides (NL)

Herbicides (Total)

Herbicides (IRE)

Herbicides (NL)

Pesticides (Total)

Pesticides (IRE)

Pesticides (NL)

DME (Total)

DME (IRE)

DME (NL)

PAH (Total)

PAH (IRE)

PAH (NL)

Gasses and fumes (Total)

Gasses and fumes (IRE)

Gasses and fumes (NL)

Nickel (Total)

Nickel (IRE)

Nickel (NL)

Chromium (Total)

Chromium (IRE)

Chromium (NL)

Metals (Total)

Metals (IRE)

Metals (NL)

Chlorinated solvents (Total)

Chlorinated solvents (IRE)

Chlorinated solvents (NL)

Aromatic solvents (Total)

Aromatic solvents (IRE)

Aromatic solvents (NL)

1.0 1.2 1.4 0.9 1.0 1.1 1.2 1.3

0.9 1.0 1.1 1.2 1.3 1.4 0.8 1.0 1.2 1.4

0.8 1.2 1.6 2.0 0.9 1.0 1.1 1.2 1.3OR

Expo

sure

*

*

*

*

*

**

Figure 5.1 ALS risk associated with cumulative job exposures in the Dutch population (NL), Irish population (IRE) and a meta-analysis of both populations (Total). Odds ratios (OR) and 95% confidence intervals are shown in the figures categorized by exposure main groups. * P value <0·05; ** P value <0·008 (Bonferroni adjusted). PAH, polycyclic aromatic hydrocarbon; DME, diesel motor exhaust.


79


DISCUSSION

Using a job exposure matrix in two independent populations, we showed that a higher cumulative lifetime exposure to DME is associated with an increased risk of ALS. No associations with ALS were found for environmental toxins of mineral dust, organic dust, pesticides, metals and solvents. Epidemiological studies objectively assessing DME exposure and the association with ALS or other neurodegenerative diseases, such as Alzheimer’s disease and Parkinson’s disease, are lacking. However, there are few occupational studies reporting on an increased ALS risk among truck drivers,27, 28 bus drivers,6 machine workers and operators,6, 29 and military personnel.7, 9 In all these occupations, people are exposed to certain amounts of DME. DME is a major component of air pollution and an important source of atmospheric particles smaller than 0·1 μm, called nanoparticles.30 To cause neurotoxicity, these nanoparticles need to be transported to the brain. Recently, animal studies showed that in mice exposed to DME the blood brain barrier is compromised, leading to an increase in neuroinflammatory markers (e.g. IgG, inducible nitric oxide synthase (iNOS), interleukin (IL)-1B) in the brain parenchym.31, 32 Moreover, there has been accumulating evidence that the olfactory nerve provides a direct route for delivery of these nanoparticles to the brain, bypassing the protective blood-brain barrier, where they may be involved in neuroinflammation and neuropathology.33, 34 Significantly higher concentrations of nanoparticles have been identified in the olfactory bulb and frontal cortex of residents in a highly polluted urban environment, compared with residents in low pollution cities.35 In these highly exposed residents, genes that are involved in inflammation were significantly upregulated. Moreover, two recent papers investigated the effect of prolonged DME exposure on the rat brain and found an increased neuroinflammatory response, with increased pro-inflammatory cytokines TNFɑ, IL-1ɑ, IL-1β.30, 36 These prior findings and our observation of an increased risk of ALS with higher levels of DME exposure, suggests a role for DME exposure in motor neuron degeneration. The toxic effect of DME may be comparable to the neurotoxicity of smoking, the only widely accepted risk factor in ALS,37 in which neuroinflammation has also been a proposed pathological mechanism.38 Performing the primary analysis without adjustment for smoking showed similar elevated effect estimates, indicating that smokers do not per se have a higher DME exposure. This suggests that both smoking and DME exposure may independently lead to neuroinflammation and subsequently to an increased risk of ALS.Based on these findings, it would be interesting to determine whether residential exposure to traffic related air pollution, as an independent method compared with the occupational exposure assessment in our study, is also associated with an increased risk of ALS.We have to acknowledge certain limitations of the present study. Due to regulations, the levels of occupational exposures in the past may be different from present levels. The JEM, however, does not take into account these changes in occupational exposures over time. Since, however, most participants in the study will have been exposed in the same period, the effect of misclassification of exposure level due to this limitation is probably small.


CHAPTER 5

80

Another limitation is that the job exposure matrix assigns a similar exposure to everyone with the same job, although exposure levels may show large variances between subjects within a job-title. A prospective cohort study with individual exposure measurements may be the only way to avoid this source of bias. However, bearing in mind the low incidence of ALS, the sample size required is so large that such a study may never be performed. The present study had some major strengths, which were the two independent population-based settings with control groups representative of the general population in each country, the good quality of data on lifetime occupational history and confounders, and application of a job exposure matrix (a valid and objective method for exposure assessment) to investigate the association of occupational exposures with ALS, avoiding the risk of recall bias or the effect of leading questions. Finally, the relative risk for DME exposure in the meta-analysis was 1·1, clearly indicating that DME exposure can be one of many, environmental and genetic, steps that are needed to develop ALS.39

PANEL RESEARCH IN CONTEXT

Systematic reviewWe searched PubMed for reports published in English before December 2014, with the following terms: (”amyotrophic lateral sclerosis”, “ALS”, “motor neuron disease”, “motor neurone disease”, or “MND”), and (“diesel motor exhaust”, “diesel exhaust”, or “DME”). We identified one study that linked occupations associated with ALS to DME exposure.27 In this study, few other epidemiological studies were suggested to report on occupations associated with ALS risk in which DME exposure may be the causal link.6, 7, 9, 27-29 None of these studies, however, used an objective and quantitative exposure assessment.

InterpretationOur study presents evidence for an increased risk of ALS with higher cumulative occupational exposure to DME in two independent population-based settings, adjusted for a wide range of potential confounder variables. In contrast with previous epidemiological studies, we used a job exposure matrix (JEM), a recognized valid, objective and agent-specific method in case-control studies, to assess occupational exposures.16-18 Here, we provide compelling evidence that exposure to DME may be one of many risk factors needed to develop ALS.


81


REFERENCES

1. Kiernan MC, et al. Amyotrophic lateral sclerosis. Lancet. 2011; 377: 942-55.2. Sutedja NA, et al. What we truly know about occupation as a risk factor for ALS: a critical and

systematic review. Amyotroph Lateral Scler. 2009; 10: 295-301.3. Belli S and Vanacore N. Proportionate mortality of Italian soccer players: is amyotrophic

lateral sclerosis an occupational disease? Eur J Epidemiol. 2005; 20: 237-42.4. Gunnarsson LG, et al. Amyotrophic lateral sclerosis in Sweden in relation to occupation. Acta

Neurol Scand. 1991; 83: 394-8.5. Nicholas JS, et al. Health among commercial airline pilots. Aviat Space Environ Med. 2001; 72:

821-6.6. Park RM, et al. Potential occupational risks for neurodegenerative diseases. Am J Ind Med.

2005; 48: 63-77.7. Haley RW. Excess incidence of ALS in young Gulf War veterans. Neurology. 2003; 61: 750-6.8. Weisskopf MG, et al. Prospective study of occupation and amyotrophic lateral sclerosis

mortality. Am J Epidemiol. 2005; 162: 1146-52.9. Weisskopf MG, et al. Prospective study of military service and mortality from ALS. Neurology.

2005; 64: 32-7.10. Weisskopf MG, et al. Prospective study of chemical exposures and amyotrophic lateral

sclerosis. J Neurol Neurosurg Psychiatry. 2009; 80: 558-61.11. Vanacore N, et al. Job strain, hypoxia and risk of amyotrophic lateral sclerosis: Results from a

deathcertificatestudy.Amyotroph Lateral Scler. 2010; 11: 430-4.12. Furby A, et al. Rural environment and risk factors of amyotrophic lateral sclerosis: a case-

control study. J Neurol. 2010; 257: 792-8.13. Fang F, et al. Workplace exposures and the risk of amyotrophic lateral sclerosis. Environ Health

Perspect. 2009; 117: 1387-92.14. McGuire V, et al. Occupational exposures and amyotrophic lateral sclerosis. A population-

based case-control study. Am J Epidemiol. 1997; 145: 1076-88.15. Kamel F, et al. Lead exposure as a risk factor for amyotrophic lateral sclerosis. Neurodegener

Dis. 2005; 2: 195-201.16. Sutedja NA, et al. Exposure to chemicals and metals and risk of amyotrophic lateral sclerosis:

a systematic review. Amyotroph Lateral Scler. 2009; 10: 302-9.17. Vergara XP, et al. New electric-shock job exposure matrix. Am J Ind Med. 2012; 55: 232-40.18. Huss A, et al. Electric shocks at work in Europe: development of a job exposure matrix. Occup

Environ Med. 2013; 70: 261-7.19. ParlettLE,etal.Evaluationofoccupationalexposuretomagneticfieldsandmotorneuron

disease mortality in a population-based cohort. J Occup Environ Med. 2011; 53: 1447-51.20. HussA,etal.OccupationalexposuretomagneticfieldsandelectricshocksandriskofALS:

The Swiss National Cohort. Amyotroph Lateral Scler Frontotemporal Degener. 2014: 1-6.21. Vergara X, et al. Case-control study of occupational exposure to electric shocks and magnetic

fieldsandmortalityfromamyotrophiclateralsclerosisintheUS,1991-1999.J Expo Sci Environ Epidemiol. 2014.

22. Sedgwick P. What is recall bias? BMJ. 2012; 344: e3519.23. Huisman MHB, et al. Population based epidemiology of amyotrophic lateral sclerosis using

capture-recapture methodology. Journal of Neurology Neurosurgery and Psychiatry. 2011; 82: 1165-70.


25. Traynor BJ, et al. Incidence and prevalence of ALS in Ireland, 1995-1997: a population-based study. Neurology. 1999; 52: 504-9.

26. International Labor Organization. International Standard Classification of Occupations.2010.

27. Pamphlett R and Rikard-Bell A. Different occupations associated with amyotrophic lateral sclerosis: is diesel exhaust the link? PLoS One. 2013; 8: e80993.


CHAPTER 5

82

28. Kurtzke JF and Beebe GW. Epidemiology of amyotrophic lateral sclerosis: 1. A case-control comparison based on ALS deaths. Neurology. 1980; 30: 453-62.

29. Schulte PA, et al. Neurodegenerative diseases: occupational occurrence and potential risk factors, 1982 through 1991. Am J Public Health. 1996; 86: 1281-8.

30. Gerlofs-Nijland ME, et al. Effect of prolonged exposure to diesel engine exhaust on proinflammatorymarkersindifferentregionsoftheratbrain.Part Fibre Toxicol. 2010; 7: 12.

31. Heidari Nejad S, et al. The effect of diesel exhaust exposure on blood-brain barrier integrity and function in a murine model. J Appl Toxicol. 2015; 35: 41-7.

32. Oppenheim HA, et al. Exposure to vehicle emissions results in altered blood brain barrier permeability and expression of matrix metalloproteinases and tight junction proteins in mice. Part Fibre Toxicol. 2013; 10: 62.

33. Lucchini RG, et al. Neurological impacts from inhalation of pollutants and the nose-brain connection. Neurotoxicology. 2012; 33: 838-41.

34. TonelliLHandPostolacheTT.Airborneinflammatoryfactors:"fromthenosetothebrain".Front Biosci (Schol Ed). 2010; 2: 135-52.

35. Calderon-GarciduenasL,etal.Neuroinflammation,hyperphosphorylatedtau,diffuseamyloidplaques, and down-regulation of the cellular prion protein in air pollution exposed children and young adults. J Alzheimers Dis. 2012; 28: 93-107.

36. Levesque S, et al. Air pollution & the brain: Subchronic diesel exhaust exposure causes neuroinflammation and elevates early markers of neurodegenerative disease. J Neuroinflammation. 2011; 8: 105.


38. Rothstein JD. Current hypotheses for the underlying biology of amyotrophic lateral sclerosis. Ann Neurol. 2009; 65 Suppl 1: S3-9.

39. Al-Chalabi A, et al. Analysis of amyotrophic lateral sclerosis as a multistep process: a population-based modelling study. Lancet Neurol. 2014; 13: 1108-13.

40. PetersS,etal.Developmentofanexposuremeasurementdatabaseonfivelungcarcinogens(ExpoSYN) for quantitative retrospective occupational exposure assessment. Ann Occup Hyg 2012; 56: 70-79.

41. Peters S, et al. Comparison of exposure assessment methods for occupational carcinogens in a multi-centre lung cancer case-control study. Occup Environ Med 2011; 68: 148-153.

42. Matheson MC, et al. Biological dust exposure in the workplace is a risk factor for chronic obstructive pulmonary disease. Thorax 2005; 60: 645-651.


83


Chromium

DME

Insecticides

Pesticides

Mineral dust

1.00 1.25 1.50 1.75 2.00Odds Ratio

Exp

osur

e

Figure S5.1 Meta-analyses of ALS risk associated with last job exposures. Meta-analyses were performed using a fixed effect model of the Dutch and Irish population data. Odds ratios (OR) and 95% confidence intervals are shown in the figure for each exposure separately. The five exposures shown, were analyzed as they all had a p value of ≤ 0·2 in the meta-analysis of cumulative exposures. DME, diesel motor exhaust.



Meinie Seelen1*, Rosario A Toro Campos2*, Jan H Veldink1, Anne E Visser1, Gerard Hoek2, Anneke J van der Kooi3, Marianne de Visser3, Joost Raaphorst4,

Leonard H van den Berg1†, Roel CH Vermeulen2†

1 Department of Neurology, Brain Center Rudolf Magnus, University Medical Center Utrecht, the Netherlands.

2 Environmental Epidemiology Division, Institute for Risk Assessment Sciences, Utrecht University, the Netherlands.

3 Department of Neurology, Amsterdam Medical Center, University of Amsterdam, the Netherlands.

4 Department of Neurology, Donders Institute for Brain, Cognition and Behaviour, Center for Neuroscience, Radboud University Nijmegen Medical Center, the Netherlands.

* These authors contributed equally† Joint last authors

In preparation

Long-term exposure to traffic related air pollution

is associated with an increased risk of

amyotrophic lateral sclerosis

CHAPTER 6


CHAPTER 6

86

SUMMARY

BackgroundLong-term exposure to air pollution has been associated with several neurodevelopmental and neurodegenerative disorders. The association with amyotrophic lateral sclerosis (ALS) has not been studied before.

MethodsWe assessed long-term exposure to multiple air pollutants and the risk of developing ALS in a large population-based case-control study in The Netherlands (917 cases and 2,662 individually matched controls recruited between 2006 and 2013). Residential annual mean air pollution concentrations, averaged over the period 1992 to enrolment, were assessed by land use regression (LUR) models developed as part of the European Study of Cohorts for Air Pollution Effects (ESCAPE). Exposure estimates included nitrogen oxides (NO2 and NOx), particulate matter (PM) with diameters of less than 2·5 μm (PM

2·5),

less than 10 μm (PM10), and between 10 μm and 2·5 μm (PMcoarse), and PM2·5absorbance (a marker for black carbon or soot). We performed conditional logistic regression analysis by quartiles of exposure and used two different multivariate models (model 1 adjusted for age, gender, education, smoking, alcohol use, body mass index, and social-economic status; model 2 additionally adjusted for urbanization rate).

FindingsRisk of ALS was significantly increased for individuals in the upper quartile of PM2·5absorbance (OR 1·57, 95% CI 1·14-2·17), NO2 (OR 1·55, 95% CI 1·08-2·21), and NOx concentrations (OR 1·38, 95% CI 1·07-1·77) when compared to the lowest exposure quartiles. These results, except for NOx, remained significant after additionally adjusting for urbanisation rate. Results were similar between smokers and non-smokers, and more pronounced for subject with a bulbar onset.

InterpretationThis is the first study to report that long-term exposure to air pollution is a potential environmental risk factor for developing ALS.


87

Exposure to air pollution and ALS risk

INTRODUCTION

Amyotrophic lateral sclerosis (ALS) is a rapidly progressive neurodegenerative disease in which motor neuron loss results in paralysis of limbs, speech and swallowing difficulties, and eventually respiratory failure. Fifty percent of patients with ALS die within three years from symptom onset.1 The lifetime risk of ALS is 1:300, occurring at any adult age, with a median age at onset of 63 years.1, 2 90-95% of ALS cases appear to be sporadic, which is thought to have a complex aetiology, most probably caused by an interaction of multiple genetic and exogenous factors.3 Smoking is thus far the only exogenous factor that has been consistently identified as a risk factor.4 Other risk factors remain inconclusive in part due to poor study design, lack of replication studies and low numbers of patients.Long-term exposure to air pollutants has been linked to increased mortality rates,5-8 specifically to cardiovascular diseases,9, 10 respiratory diseases,7, 11, 12 and to a lesser extent to neurodevelopmental and neurodegenerative diseases including autism,13 Parkinson’s14 and Alzheimer’s disease.15 To date, there are no epidemiological studies on the risk of developing ALS and air pollution. However, a consistent association between smoking and the development of ALS has been observed in both case-control and cohort studies.4 As in smoking, exposure to air pollutants, in particular to fine particulate matter, have been hypothesised to be able to cross the blood brain barrier, or impair the blood brain-barrier leading to inflammatory and oxidative stress responses in the brain.16

We investigated the association between multiple air pollutants and the risk of ALS by using historic residential data from a large population-based case-control study on ALS and exposure data from the European Study of Cohorts for Air Pollution Effects (ESCAPE) project.

METHODS

Study populationALS patients diagnosed between January 2006 and January 2013 in The Netherlands were enrolled into the Prospective ALS study the Netherlands (PAN). The PAN is a large population-based case-control study with an estimated capture rate of 81% of all ALS cases in The Netherlands.1 All patients newly diagnosed as possible, probable (laboratory supported) or definite ALS according to the revised El Escorial Criteria were included.17 Excluded were ALS mimics and patients who had a first, second or third degree family member with ALS, defined as familial ALS. Controls were selected through the general practitioner of the patient, which is a representative tool to select population-based controls. For the purpose of this study controls were post-hoc matched to the cases by gender, age (+/- 5 years), region of residence and enrolment date (+/-1 year). Spouses or blood-relatives of patients were not eligible to be controls to prevent overmatching. The institutional review board of the University Medical Center Utrecht provided ethical approval. All participants gave written informed consent for inclusion in the study.


CHAPTER 6

88

Exposure assessmentWe estimated long-term exposure to air pollutants at the residential address of the study participants from 1992 to the date of enrollment in the study using recently developed Land Use Regression (LUR) models within the ESCAPE project.18, 19 In brief, air pollution was repeatedly measured at multiple locations in 2009 to derive annual average concentration of NO2, NOx, PM2·5 (fraction of PM smaller than 2.5 μm), PM10

(fraction of PM smaller than 10 μm), PMcoarse (fraction of PM calculated as the concentration of PM10 minus that of PM2.5), and PM2·5 absorbance (marker for soot or black carbon). Subsequently, LUR models were developed to explain the spatial variability in air pollutants by spatial indicators such as traffic intensity, population density and land use. These LUR models were then used to estimate annual ambient air pollution concentration at the participants’ addresses. To allow for variation in air pollution concentrations overtime, as case-control recruitment varied between 2006 and 2013, we back extrapolated modeled concentrations in 2009 to 1992. We did not back-extrapolate further since this was the first year in which routine monitoring data was available for all pollutants in The Netherlands. Constant concentrations were assumed for the period 2009-2013. Subsequently, we averaged the annual average air pollutant concentrations for each individual from 1992 to the date of enrollment in the study. Participants (22 cases, 15 controls), of whom more than 50% of the addresses between 1992 until onset or inclusion were missing, were excluded. We also performed analyses without any historical back extrapolation or by using the air pollution estimate in 1992 for the whole population. These analyses did not result in marked differences as compared to the original exposure assignment and are therefore not presented.

Statistical analysisAverage annual air pollution exposure was divided in quartiles based on the exposure distribution among the controls. Conditional logistic regression models were used to determine the association between exposure to air pollutants and ALS. In addition, p values for linear trend were calculated using the median value in each quartile as a continuous variable. We specified two a-priori models to adjust for confounding based on known and putative risk factors of ALS. Model 1 was adjusted for age, sex, education (three levels: elementary school, middle/high school and college/university), premorbid body mass index (BMI), current (before disease onset for cases) smoking status, current alcohol use and area social economic status (SES; percentage high income at the municipality level of residency). In model 2 we added urbanization rate as a potential confounder to allow for a stronger control on urban, peri-urban and rural differences in lifestyle and environmental factors. Missing values of confounder variables were imputed with the R package Hmisc using multiple reiterations (n=10) of predictive mean matching with optional weighted probability sampling of the other variables.We performed several sensitivity analyses. First, we explored the possible effect of recency of exposure by only counting the last year of exposure before onset of ALS, or the last five years before onset. Secondly, we considered a one and five year lag prior to symptom


89


onset of ALS, to determine whether incipient ALS was of influence on the association. Thirdly, we restricted the analyses to participants who did not move during the last year or the last five years before symptom onset, to exclude reverse causation of patients moving closer to academic treatment centers, which are located in the larger cities in The Netherlands. Fourth, we restricted the analysis to the cases and controls with all confounder data available, excluding an effect of imputation. Lastly, we assessed possible effect modification of smoking status, and we performed a subgroup analyses by site of symptom onset (patients with bulbar or spinal onset).

Role of the funding sourceThe sponsor of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. MS, RT, LHB and RV had full access to all the data in the study. LHB and RV had final responsibility for the decision to submit for publication.

RESULTS

The presented analyses are based on 917 patients with ALS and 2,662 individually matched controls. Clinical characteristics of the patients with ALS, such as age at onset, site of onset and El Escorial classification, were similar to previously reported patient characteristics in Europe (Table 6.1).20 BMI, current smoking, current alcohol use, area SES and urbanization rate were significantly different between patients and controls (Table 6.1). Data on at least one of the confounder variables education, BMI, smoking and alcohol use were missing in 27·5% of the participants and subsequently imputed. Sensitivity analysis restricted to the non-imputed population did not result in substantially different results compared to the analysis of the imputed population.The mean annual concentration for each pollutant is presented in Table 6.2 by case control status. Although, differences were small, the mean concentrations were significantly higher among the cases than controls for PM10, PMcoarse, PM2·5 absorbance, NO2, and NOx (p<0·05, Mann-Withney U). Pearson correlations between the different exposure measures were generally higher than 0·6, except for PM2·5.The ORs of all air pollutants were elevated in the category with the highest exposed individuals compared to the reference category with the lowest exposed individuals (Table 6.3). For PM2·5absorbance, NO2, and NOx, these ORs were significantly elevated in model 1 (OR 1·67 (95% CI 1·27-2·18), OR 1·74 (95% CI 1·32-2·30), and OR 1·38 (95% CI 1·07-1·77)). A slight decrease in the association was found between model 1 adjusted for age, sex, education, BMI, smoking, alcohol, and area SES and the more comprehensive adjustment model 2 (with inclusion of urbanization rate). In model 2, PM2·5absorbance and NO2, still showed significantly elevated ORs in the highest exposure category (OR 1·57 (95% CI 1·14-2·17) and OR 1·55 (95% CI 1·08-2·21)). Sensitivity analyses by recency of exposure, by lagging exposures, or by limiting the analyses to people that did not move homes in the last years did not result in materially


CHAPTER 6

90

different results. Stratification by smoking status revealed that the ORs for never smokers were similar to the overall ORs (Table S6.1). Patients with a bulbar site of onset showed higher increased ALS risks with exposure to air pollutants compared to patients with a spinal onset (Table S6.2).



(n=917) (n=2662) p value

Male, n (%) 560 (61·1) 1633 (61·3) 0·88

Age, y, median (IQR)* 63·5 (57·0-70·1) 63·5 (57·5-69·7) 0·81

Bulbar site of onset, n (%) 321 (35·0)


Definite 166 (18·1)

Probable 377 (41·1)

Probable lab supported 215 (23·4)

Possible 144 (15·7)

Education, n (%)

Elementary school 74 (8·1) 190 (7·1) 0·23

Middle school/High school 603 (65·8) 1702 (63·9)

College/University 240 (26·2) 770 (28·9)

Premorbid BMI, kg/m2, n (%)

Underweight (<18·5) 30 (3·3) 23 (0·9) <0·001

Normal weight (18·5 - <25·0) 512 (55·8) 1143 (42·9)

Overweight (25·0 - <30·0) 300 (32·7) 1239 (46·5)

Obese (≥30·0) 75 (8·2) 257 (9·7)

Current smoking, n (%)§ 154 (16·8) 346 (13·0) 0·004

Current alcohol consumption, n (%)§ 699 (76·2) 2283 (85·8) <0·001

Area SES, median (IQR)† 20·0 (18·0-23·0) 20·6 (18·0-24·0) 0·03

Urbanization rate, addresses/km2, n (%)‡

Very high (≥2500) 121 (13·2) 243 (9·1) <0·001

High (1500 - <2500) 244 (26·6) 683 (25·7)

Moderately high (1000 - <1500) 235 (25·6) 659 (24·8)

Low (500 - <1000) 245 (26·7) 755 (28·4)

Very low (<500) 72 (7·9) 322 (12·1)

ALS = amyotrophic lateral sclerosis; BMI = body mass index; SES = social economic status. *Age at onset in patients, and age on which questionnaire was completed in controls. §Current = before onset of ALS. †Social economic status is based on area level, percentage of high income is depicted. ‡Urbanization rate is based on area level, divided in five categories according to the Statistics Netherlands (www.cbs.nl).


91


DISCUSSION

In this study, we observed an increased risk of ALS associated with long-term exposure to air pollution, specifically PM2·5absorbance and the nitrogen oxides. Associations remained unchanged over the different multivariate models, except for NOx, and in sensitivity analyses. This study is the first to describe the effects of long-term residential air pollution exposure and ALS risk. Of the different air pollutants studied here, PM2·5absorbance, NO2 and NOx are primary traffic related pollutants, and therefore have larger spatial concentration differences in urban areas. On the contrary, inhalable, coarse and fine particles (PM10, PMcoarse, and PM2·5) have a considerable contribution from secondary road dust, agricultural, and construction industries. In this study, we specifically found an increased ALS risk for the more traffic related air pollutants PM2·5absorbance, NO2 and NOx. Interestingly, diesel exhaust is an important source of traffic related air pollution, and this exposure has previously been associated with ALS risk in several occupational studies: elevated ALS risk was reported for truck drivers,21, 22 bus drivers,23 and machine workers and operators.23, 24

The observation parallels also the observations with smoking, as the only established environmental risk factor in ALS.4 Smoking increases the risk of ALS through potentially several mechanisms, including inflammation, oxidative stress and direct neurotoxicity caused by fine particles, heavy metals and other chemical compounds present in cigarette smoke.25, 26 Long-term exposure to air pollution produces inflammatory damage to the cardiopulmonary system, but it may also lead to chronic inflammation and oxidative stress in the brain.27, 28 Previous studies demonstrated that ultrafine particles can circumvent the blood-brain barrier by deposition on the olfactory mucosa of the nasal region.27-29 The particles are then translocated along the olfactory nerve into the olfactory bulb of the brain, and may travel transneuronally to more distal sites within the brain.30,

31 In an autopsy study, residents in highly polluted urban areas had significantly higher concentrations of fine particulates in the olfactory bulb and frontal cortex compared to residents in low polluted areas.32 Recent experimental studies showed that there is also

Table 6.2 Mean annual air pollution exposure for ALS patients and controls


(n=917) (n=2662) p value

PM10, μg/m3, mean (SD) 31·8 (1·4) 31·6 (1·2) <0·001

PMcoarse, μg/m3, mean (SD) 10·4 (0·7) 10·3 (0·6) <0·001

PM2·5, μg/m3, mean (SD) 21·3 (0·9) 21·2 (0·8) 0·08

PM2·5 absorbance, 10-5 m-1, mean (SD) 1·43 (0·20) 1·40 (0·18) <0·001

NO2, μg/m3, mean (SD) 26·9 (5·6) 26·0 (5·2) <0·001

NOx, μg/m3, mean (SD) 44·7 (9·6) 43·6 (8·9) 0·002

ALS = amyotrophic lateral sclerosis.


CHAPTER 6

92

Table 6.3 The association between ALS risk and exposure to air pollution in two different multivariate models

Model 1* Trend Model 2* Trend

OR (95% CI) p value OR (95% CI) p Value

PM10 (μg/m3)

Q1 (≤30·9) Reference 0·006 Reference 0·19

Q2 (>30·9 - ≤31·6) 0·77 (0·59-1·00) 0·75 (0·57-0·98)

Q3 (>31·6 - ≤32·2) 0·83 (0·62-1·10) 0·77 (0·57-1·05)

Q4 (>32·2) 1·29 (0·97-1·72) 1·12 (0·79-1·57)

PMcoarse (μg/m3)


Q2 (>9·9 - ≤10·2) 0·82 (0·64-1·05) 0·77 (0·60-1·00)

Q3 (>10·2 - ≤10·5) 0·95 (0·74-1·24) 0·84 (0·64-1·11)

Q4 (>10·5) 1·24 (0·95-1·61) 1·04 (0·77-1·41)

PM2·5 (μg/m3)


Q2 (>20·7 - ≤21·3) 0·98 (0·75-1·28) 0·92 (0·70-1·21)

Q3 (>21·3 - ≤21·7) 0·85 (0·62-1·16) 0·80 (0·59-1·09)

Q4 (>21·7) 1·35 (0·97-1·88) 1·24 (0·89-1·73)

PM2·5 absorbance (10-5 m-1)

Q1 (≤1·29) Reference <0·001 Reference 0·002

Q2 (>1·29 - ≤1·38) 1·14 (0·88-1·47) 1·11 (0·85-1·44)

Q3 (>1·38 - ≤1·49) 1·12 (0·86-1·47) 1·09 (0·81-1·47)

Q4 (>1·49) 1·67 (1·27-2·18) 1·57 (1·14-2·17)

NO2 (μg/m3)

Q1 (≤22·5) Reference <0·001 Reference 0·03

Q2 (>22·5 - ≤25·8) 1·38 (1·09-1·76) 1·29 (1·01-1·66)

Q3 (>25·8 - ≤29·0) 1·25 (0·97-1·63) 1·15 (0·85-1·55)

Q4 (>29·0) 1·74 (1·32-2·30) 1·55 (1·08-2·21)

NOx (μg/m3)


Q2 (>38·2 - ≤42·2) 0·98 (0·78-1·24) 0·91 (0·71-1·15)

Q3 (>42·2 - ≤47·3) 1·12 (0·87-1·43) 0·99 (0·76-1·30)

Q4 (>47·3) 1·38 (1·07-1·77) 1·17 (0·87-1·57)

*Model 1 was adjusted for sex, age, educational level, body mass index, current smoking, current alcohol consumption, and area-level socioeconomic status; model 2 was adjusted as in model 1, but also for urbanization rate.


93


another route for small particles to enter the brain: mice exposed to diesel exhaust had a compromised blood brain barrier, leading to an increase in neuroinflammatory markers in the brain.33, 34 Moreover, histological evidence showed that human and animal brains exposed to high PM concentrations had increased levels of pro-inflammatory cytokines and markers of oxidative stress (e.g. TNF-ɑ, interleukins, NF-κB, Toll-like receptor).31, 32, 35-37

As smoking is a known risk factor for ALS we performed stratified analyses by smoking status. Results among non-smokers and former and current smokers were essentially similar indicating that the observed effect of air pollution is not easily explained by confounding due to smoking nor that there is evidence of effect modification by smoking. Interestingly, we did observe a difference in effect estimates for the subgroups of site of symptom onset, with stronger associations for patients with a bulbar onset. This association has not been reported for smoking and it is unknown how exposure to air pollutants may favour a bulbar site of onset. The only, speculative, reason could be that the bulbar region is physically closer to the olfactory region, as compared to the spinal lower motor neurons. Since it’s hypothesized that ALS “spreads” through the CNS by a prion-like mechanism after the initial trigger, this could explain the more frequent bulbar site of onset.The most important limitation in our study is the uncertainty in air pollution estimates. However, land use regression models have previously been reported to predict the historic spatial variation very well and the ESCAPE models used in this study have been shown to detect known risk of air pollution well.38, 39 Nevertheless, noteworthy limitations in the exposure assessment are that we had only data on air pollution exposure from 1992 onwards. The early years of exposure (before 1992) might have been relevant in ALS pathogenesis as well. However, previous studies have shown that air pollution assessment using the baseline address versus a complete residential history only leads to minimal attenuation of effect parameters.38, 40 That lack of data before 1992 may potentially have resulted in some non-differential misclassification resulting most likely in bias towards the null. This study provides new clues for pathogenic pathways in ALS, ultimately leading towards a better understanding, and prevention strategies. As it is the first study to report on this possible association it is important for other studies to replicate the findings. Second to that, our novel findings of an increased risk of developing ALS are with exposure levels that ranges well below the existing European annual mean limit values of 25 μg/m3 for PM2·5, and 40 μg/m3 (for PM10 and NO2) (European Commission air quality standards).41 As ambient air pollution levels are modifiable, this adds to the necessity of regulatory public health interventions on air pollution exposure levels.


CHAPTER 6

94

PANEL RESEARCH IN CONTEXT

Systematic reviewWe searched PubMed for reports published in English before November, 2014, with the following terms: (”amyotrophic lateral sclerosis”, “ALS”, “motor neuron disease”, “motor neurone disease”, or “MND”), and (“air pollution”, “air pollutants”, or “particulate matter”). We identified no previous epidemiological studies assessing residential exposure to particulate matter air pollution in ALS. We used a method previously described within the multicenter ESCAPE project to assess the effect of air pollution on mortality, and cardiopulmonary diseases6, 9, 11 and applied this method to amyotrophic lateral sclerosis population data in the Netherlands.

InterpretationOur study has a large, population-representative, sample size, with a highly standardized exposure assessment, and contains adjustment for a wide range of potential confounder variables. Here, we provide evidence for the association between long-term exposure to traffic related air pollution and an increased risk of developing ALS. Air pollution may be one of the environmental risk factors of ALS that is ubiquitous and can be modified at a population level. Therefore, future research replicating our findings would be of great importance, even for a less common disorder such as ALS.


95


REFERENCES


2. Cronin S, et al. Ethnic variation in the incidence of ALS: a systematic review. Neurology. 2007; 68: 1002-7.

3. Al-Chalabi A and Hardiman O. The epidemiology of ALS: a conspiracy of genes, environment and time. Nat Rev Neurol. 2013; 9: 617-28.


5. Dockery. An association between air pollution and mortality in six U.S. cities. 1993.6. Beelen R, et al. Effects of long-term exposure to air pollution on natural-cause mortality: an

analysis of 22 European cohorts within the multicentre ESCAPE project. Lancet. 2014; 383: 785-95.

7. BeelenR,etal.Long-termeffectsoftraffic-relatedairpollutiononmortalityinaDutchcohort(NLCS-AIR study). Environ Health Perspect. 2008; 116: 196-202.

8. Cesaroni G, et al. Long-term exposure to urban air pollution and mortality in a cohort of more than a million adults in Rome. Environ Health Perspect. 2013; 121: 324-31.

9. Cesaroni G, et al. Long term exposure to ambient air pollution and incidence of acute coronary events: prospective cohort study and meta-analysis in 11 European cohorts from the ESCAPE Project. BMJ. 2014; 348: f7412.

10. Raaschou-NielsenO,etal.Trafficairpollutionandmortalityfromcardiovasculardiseaseandall causes: a Danish cohort study. Environ Health. 2012; 11: 60.

11. Dimakopoulou K, et al. Air pollution and nonmalignant respiratory mortality in 16 cohorts within the ESCAPE project. Am J Respir Crit Care Med. 2014; 189: 684-96.

12. Dong GH, et al. Long-term exposure to ambient air pollution and respiratory disease mortality in Shenyang, China: a 12-year population-based retrospective cohort study. Respiration. 2012; 84: 360-8.

13. VolkHE,etal.Traffic-relatedairpollution,particulatematter,andautism.JAMA Psychiatry. 2013; 70: 71-7.

14. Finkelstein MM and Jerrett M. A study of the relationships between Parkinson's disease and markersoftraffic-derivedandenvironmentalmanganeseairpollutionintwoCanadiancities.Environ Res. 2007; 104: 420-32.

15. Ranft U, et al. Long-term exposure to traffic-related particulate matter impairs cognitivefunction in the elderly. Environ Res. 2009; 109: 1004-11.

16. Costa LG, et al. Neurotoxicants are in the air: convergence of human, animal, and in vitro studies on the effects of air pollution on the brain. Biomed Res Int. 2014; 2014: 736385.


18. Eeftens M, et al. Development of Land Use Regression models for PM(2.5), PM(2.5) absorbance, PM(10) and PM(coarse) in 20 European study areas; results of the ESCAPE project. Environ Sci Technol. 2012; 46: 11195-205.

19. Beelen R, et al. Development of NO2 and NOx land use regression models for estimating air pollution exposure in 36 study areas in Europe - The ESCAPE project. Atmospheric Environment. 2013; 72: 10-23.

20. Logroscino G, et al. Incidence of amyotrophic lateral sclerosis in Europe. J Neurol Neurosurg Psychiatry. 2010; 81: 385-90.

21. Pamphlett R and Rikard-Bell A. Different occupations associated with amyotrophic lateral sclerosis: is diesel exhaust the link? PLoS One. 2013; 8: e80993.

22. Kurtzke JF and Beebe GW. Epidemiology of amyotrophic lateral sclerosis: 1. A case-control comparison based on ALS deaths. Neurology. 1980; 30: 453-62.

23. Park RM, et al. Potential occupational risks for neurodegenerative diseases. Am J Ind Med. 2005; 48: 63-77.


CHAPTER 6

96

24. Schulte PA, et al. Neurodegenerative diseases: occupational occurrence and potential risk factors, 1982 through 1991. Am J Public Health. 1996; 86: 1281-8.


26. Alonso A, et al. Association of smoking with amyotrophic lateral sclerosis risk and survival in men and women: a prospective study. BMC Neurol. 2010; 10: 6.

27. TonelliLHandPostolacheTT.Airborneinflammatoryfactors:"fromthenosetothebrain".Front Biosci (Schol Ed). 2010; 2: 135-52.


29. OberdorsterG,etal.Translocationof inhaledultrafineparticles to thebrain. Inhal Toxicol. 2004; 16: 437-45.

30. Tjalve H and Henriksson J. Uptake of metals in the brain via olfactory pathways. Neurotoxicology. 1999; 20: 181-95.

31. Elder A, et al. Translocation of inhaled ultrafinemanganese oxide particles to the centralnervous system. Environ Health Perspect. 2006; 114: 1172-8.

32. Calderon-GarciduenasL,etal.Neuroinflammation,hyperphosphorylatedtau,diffuseamyloidplaques, and down-regulation of the cellular prion protein in air pollution exposed children and young adults. J Alzheimers Dis. 2012; 28: 93-107.

33. Heidari Nejad S, et al. The effect of diesel exhaust exposure on blood-brain barrier integrity and function in a murine model. J Appl Toxicol. 2015; 35: 41-7.

34. Oppenheim HA, et al. Exposure to vehicle emissions results in altered blood brain barrier permeability and expression of matrix metalloproteinases and tight junction proteins in mice. Part Fibre Toxicol. 2013; 10: 62.

35. Gillespie P, et al. Particulate matter neurotoxicity in culture is size-dependent. Neurotoxicology. 2013; 36: 112-7.

36. PetersA,etal.Translocationandpotentialneurologicaleffectsoffineandultrafineparticlesacritical update. Part Fibre Toxicol. 2006; 3: 13.

37. Levesque S, et al. Air pollution & the brain: Subchronic diesel exhaust exposure causes neuroinflammation and elevates early markers of neurodegenerative disease. J Neuroinflammation. 2011; 8: 105.

38. Eeftens M, et al. Stability of measured and modelled spatial contrasts in NO(2) over time. Occup Environ Med. 2011; 68: 765-70.

39. Gulliver J, et al. Development and back-extrapolation of NO2 land use regression models for historic exposure assessment in Great Britain. Environ Sci Technol. 2013; 47: 7804-11.

40. Cesaroni G, et al. Nitrogen dioxide levels estimated from land use regression models several years apart and association with mortality in a large cohort study. Environ Health. 2012; 11: 48.

41. European Commission. European Commission Air Quality Standards. 2014.


97



Table S6.1 Subgroup analysis of non-smokers for the association between ALS and exposure to air pollution

Non-smokers*

OR (95% CI)† p value‡

PM10

Q1 Reference 0·18

Q2 0·72 (0·54-0·97)

Q3 0·83 (0·59-1·17)

Q4 1·15 (0·78-1·69)

PMcoarse

Q1 Reference 0·06

Q2 0·78 (0·59-1·03)

Q3 0·92 (0·67-1·25)

Q4 1·19 (0·85-1·67)

PM2·5

Q1 Reference 0·15

Q2 1·01 (0·74-1·39)

Q3 0·95 (0·66-1·35)

Q4 1·34 (0·91-1·97)

PM2·5 absorbance

Q1 Reference 0·01

Q2 1·01 (0·76-1·35)

Q3 1·06 (0·76-1·48)

Q4 1·48 (1·03-2·14)

NO2

Q1 Reference 0·09

Q2 1·31 (0·99-1·74)

Q3 1·21 (0·87-1·68)

Q4 1·48 (1·00-2·20)

NOx

Q1 Reference 0·16

Q2 0·93 (0·71-1·22)

Q3 1·07 (0·79-1·44)

Q4 1·20 (0·86-1·66)

*Non-smokers: no. of cases=759, no. of controls=1956. †Main model 2 was used for the comparison; this confounder model was adjusted for sex, age, educational level, current smoking, current alcohol consumption, body mass index and socioeconomic status. ‡P value was calculated for linear trend.


CHAPTER 6

98

Table S6.2 Subgroup analyses for site of onset for the association between ALS and exposure to air pollution

Bulbar site of onset* Spinal site of onset*

OR (95% CI)† p value‡ OR (95% CI)† p value‡

PM10

Q1 Reference 0·10 Reference 0·65

Q2 0·77 (0·48-1·23) 0·72 (0·52-1·00)

Q3 0·81 (0·47-1·41) 0·76 (0·52-1·10)

Q4 1·41 (0·76-2·61) 1·00 (0·66-1·51)

PMcoarse


Q2 0·86 (0·56-1·33) 0·72 (0·52-0·98)

Q3 1·13 (0·70-1·84) 0·73 (0·52-1·02)

Q4 1·36 (0·79-2·34) 0·94 (0·65-1·36)

PM2·5


Q2 1·05 (0·66-1·67) 0·87 (0·62-1·22)

Q3 0·87 (0·50-1·49) 0·78 (0·53-1·15)

Q4 1·19 (0·67-2·13) 1·28 (0·84-1·93)

PM2·5 absorbance


Q2 1·17 (0·73-1·87) 1·08 (0·79-1·48)

Q3 1·22 (0·72-2·07) 1·02 (0·72-1·47)

Q4 2·07 (1·17-3·68) 1·36 (0·92-2·02)

NO2

Q1 Reference <0·001 Reference 0·90

Q2 1·19 (0·76-1·87) 1·33 (0·98-1·18)

Q3 1·77 (1·08-2·93) 0·91 (0·63-1·33)

Q4 2·85 (1·53-5·33) 1·15 (0·74-1·79)

NOx


Q2 0·98 (0·63-1·52) 0·90 (0·67-1·21)

Q3 1·51 (0·94-2·42) 0·81 (0·58-1·13)

Q4 1·61 (0·97-2·68) 1·01 (0·70-1·45)

*Subgroups: bulbar site of onset (no. of cases=324, no. of controls=940), spinal site of onset (no. of cases=596, no. of controls=1731). †Main model 2 was used for the comparison; this confounder model was adjusted for sex, age, educational level, current smoking, current alcohol consumption, body mass index and socioeconomic status. ‡P value was calculated for linear trend.


99



Meinie Seelen1, Roel C.H. Vermeulen2, Levien S. van Dillen1, Anneke J. van der Kooi3,Anke Huss2, Marianne de Visser3, Leonard H. van den Berg1*, Jan H. Veldink1*


2 Institute for Risk Assessment Science, Division of Epidemiology, Utrecht University, The Netherlands

3 Department of Neurology, Amsterdam Medical Center, University of Amsterdam, The Netherlands

* These authors contributed equally

Published in: Neurology (2014)

CHAPTER 7

Residential exposure to extremely low frequency

electromagnetic fields and the risk of ALS


CHAPTER 7

102

INTRODUCTION

Several studies have reported on the possible association between the risk of developing amyotrophic lateral sclerosis (ALS) and employment in the electrical industry, which may be related to extremely low frequency electromagnetic field (ELF-EMF) exposure and/or the risk of experiencing an electric shock, although no direct association has been proven.1 Three previous studies reported on ALS risk related to living near power lines, an important source of ELF-EMF exposure for the general population.2-4 These studies reported a null finding but had some shortcomings as they were based on registry data and had no detailed clinical data available. We, therefore, performed a large population-based case-control study with detailed phenotypic data, to assess the relation between residential exposure to ELF-EMF from power lines and the risk of ALS.

MethodsWe included a total of 1139 ALS patients and 2864 frequency-matched controls, derived from a large population-based case-control study performed in the Netherlands from January 2006 to January 2013.5 Controls were selected from the roster of the general practitioner of the patient and were matched to the patient on gender and age. The Netherlands has an extensive network of overhead power lines, the locations of which are known exactly.6 These power lines were classified into high voltage (50 kV, 110 kV and 150 kV) and very high voltage (220 kV and 380 kV) lines. We collected data on lifetime residential history from the Municipal Personal Records Database. Residential data of ALS patients and controls were geocoded (assigned coordinates to the addresses) taking the time of onset of disease into account.

Table 7.1 Baseline characteristics of patients and controls


(n=1139) (n=2864)

Male gender, n (%) 680 (59.7) 1772 (61.9)

Age, median (range)a 63.5 (22.4-88.5) 63.5 (20.2-91.9)

Site of onset, n (%)

Bulbar 398 (34.9)

Spinal 736 (64.6)

El Escorial Classification, n (%)

Definite 212 (18.6)

Probable 479 (42.1)


Possible 189 (16.6)

Abbreviations: ALS = amyotrophic lateral sclerosis. aAge at symptom onset for patients; age at inclusion for controls.


103

Exposure to electromagnetic fields and ALS risk

Since distance to power lines is closely associated with ELF-EMF exposure, we categorized places of residence into corridors of distance (0-<50 m, 50-<200 m, 200-<600 m and ≥600 m). We determined the shortest distance to any power line before onset of disease, and computed the cumulative years lived within 100 m of a power line, divided into 4 categories (<5 y, 5-<10 y, 10-<15 y, ≥15 y), to detect a possible dose-response relation. Statistical analyses were performed using logistic regression for the association with ALS, adjusted for gender and age (age at onset for patients and age at inclusion for controls). Subsequently, we used cox regression for the association with survival (adjusted for gender, age at onset and site of onset) and for the association with age at onset (adjusted for gender and site of onset), with the exposure variable dichotomized into living <200 m and ≥200 m from a power line.

RESULTS

Baseline characteristics of patients and controls are shown in Table 7.1. We found no increased risk of ALS in persons living in close vicinity of a power line compared to persons who had always lived at a distance of at least 600 m (Table 7.2). Cumulative exposure in years showed no dose-response relationship (data not shown). Survival analysis in ALS patients showed a non-significant hazard ratio (HR) of 1.27 (95% CI 0.87-1.86, p=0.22). Nor was there an association between distance to a power line and age at onset (HR 1.22, 95% CI 0.89-1.67, p=0.22).

Table 7.2 Frequencies and odds ratios of ALS patients and controls in relation to the shortest distance ever lived to a power line


Distance (m) n (%) n (%) OR 95% CI p-value

Very high voltage (220-380 kV)

0-<50 0 (0.0) 1 (0.0) - - -

50-<200 2 (0.2) 7 (0.2) 0.73 0.15-3.50 0.69

200-<600 23 (2.0) 44 (1.5) 1.31 0.79-2.18 0.30

≥600 1114 (97.8) 2812 (98.2) 1.00 Reference -

High voltage (50-150 kV)

0-<50 6 (0.5) 14 (0.5) 1.05 0.40-2.75 0.92

50-<200 32 (2.8) 88 (3.1) 0.91 0.60-1.37 0.64

200-<600 90 (7.9) 249 (8.7) 0.89 0.69-1.14 0.36

≥600 1011 (88.8) 2513 (87.7) 1.00 Reference -

Abbreviations: ALS = amyotrophic lateral sclerosis; m = meters; n = number; OR = odds ratio; CI = confidence interval; kV = kilovolt. Odd ratios were computed by logistic regression adjusting for gender and age.


CHAPTER 7

104

DISCUSSION

Subsequent to our null findings, we performed a meta-analysis combining our results with two previously published case-control studies.3, 4 A fixed effect model showed an overall odds ratio of 0.90 (95% CI 0.73-1.10) for subjects living <200 m compared to ≥200 m from any high voltage power line (Figure S7.1). One of these studies only assessed the current address at time of death,4 so important historical residential data regarding ELF-EMF exposure before onset of disease might be missing. A third cohort study substantiated our negative results, reporting a HR of 0.88 (95% CI 0.47-1.64).2

We did not find an association of ALS with residential exposure to ELF-EMF. This is consistent with the lack of such an association in a previously published meta-analysis in electrical occupations.1 Strengths of this study are the population-based study design, inclusion of large numbers of patients and age and gender matched controls and prevention of recall bias by the use of the Municipal Personal Records Database for collection of residential data. A limitation of this study may be the low number of participants living in close vicinity to power lines (<200 m). However, taking all studies together, one can conclude that exposure to ELF-EMF from power lines does not increase the risk of developing ALS.


105

Exposure to electromagnetic fields and ALS risk

REFERENCES

1. Vergara X, et al. Occupational exposure to extremely low-frequency magnetic fields andneurodegenerative disease: a meta-analysis. J Occup Environ Med. 2013; 55: 135-46.

2. Huss A, et al. Residence near power lines and mortality from neurodegenerative diseases: longitudinal study of the Swiss population. Am J Epidemiol. 2009; 169: 167-75.


4. Marcilio I, et al. Adult mortality from leukemia, brain cancer, amyotrophic lateral sclerosis andmagneticfieldsfrompowerlines:acase-controlstudyinBrazil.Rev Bras Epidemiol. 2011; 14: 580-8.


6. TenneT. Netkaart Nederland. 2011.


Study

Fixed effect modelHeterogeneity: I−squared=0%

Marcilio, 2011 [BRA]Frei, 2013 [DEN]Seelen, 2014 [NL]

Patients

36729901139

Controls

4706149962864

OR

0.90

0.700.980.92

[95%CI]

[0.73; 1.10]

[0.45; 1.09][0.73; 1.33][0.64; 1.33]

0.3 0.5 1 2

OR [95% CI]

Figure S7.1 The overall odds ratio of a fixed effect model of two previously published studies combined with our study in a forest plot, showing a non-significant result for subjects living <200m compared to ≥200m from a power line



M. Seelena, A.E. Vissera, D.J. Overstea, H.J. Kimb, A. Paludb, T.H. Wongc, J.C. van Swietenc,d, P. Scheltensd, N.C. Voermanse, F. Baasf, J.M.B.V. de Jongg,

A.J. van der Kooig, M. de Visserg, J.H. Veldinka, J. Paul Taylorb, M.A. Van Esa*, L.H. van den Berga*

a Department of Neurology, Brain Center Rudolf Magnus, University Medical Center Utrecht, The Netherlands

b Department of Developmental Neurobiology, St Jude Children’s Research Hospital, USAc Department of Neurology, Erasmus Medical Center, Rotterdam, The Netherlands

d Department of Neurology, VU University Medical Center, Amsterdam, The Netherlandse Department of Neurology, Donders Institute for Brain, Cognition and Behavior,

Center for Neurosciences, Radboud University Nijmegen Medical Center, The Netherlands f Department of Genome Analysis, Amsterdam Medical Center,

University of Amsterdam, The Netherlandsg Department of Neurology, Amsterdam Medical Center,

University of Amsterdam, The Netherlands


Published in: Neurobiology of Aging (2014)

No mutations in hnRNPA1 and hnRNPA2B1 in

Dutch patients with amyotrophic lateral sclerosis,

frontotemporal dementia and inclusion body myositis

CHAPTER 8

PART III - GENETICS


CHAPTER 8

108

ABSTRACT

Inclusion body myopathy (IBM) associated with Paget’s disease of the bone (PDB), fronto-temporal dementia (FTD) and amyotrophic lateral sclerosis (ALS), sometimes called IBMPFD/ALS or multi system proteinopathy (MSP), is a rare, autosomal dominant disorder characterized by progressive degeneration of muscle, brain, motor neurons and bone with prominent TDP-43 pathology. Recently two novel genes for MSP were discovered; heterogeneous nuclear ribonucleoprotein (hnRNP) A1 and A2B1. Subsequently, a mutation in hnRNPA1 was also identified in a pedigree with autosomal dominant familial ALS. The genetic evidence for ALS and other neurodegenerative diseases is still insufficient. We therefore sequenced the prion like domain of these genes in 135 familial ALS, 1084 sporadic ALS, 68 familial FTD, 74 sporadic FTD and 31 sporadic IBM patients in a Dutch population. We did not identify any mutations in these genes in our cohorts. Mutations in hnRNPA1 and hnRNPA2B1 prove to be a rare cause of ALS, FTD and IBM in the Netherlands.

INTRODUCTION

Multi system proteinopathy (MSP) is a rare, autosomal dominant, degenerative disorder, which affects multiple organ systems (brain, spinal cord, bone and muscle). Patients may present with inclusion body myopathy (IBM), frontotemporal dementia (FTD), amyotrophic lateral sclerosis (ALS) or Paget’s disease of the bone (PDB). Patients with MSP can be clinically indistinguishable from patients affected by the familial or sporadic forms of these diseases. The affected organs in MSP show prominent TDP-43 pathology, which is shared feature with ALS and FTD.1, 2 About 50% of MSP cases has been linked to mutations in the valosin-containing protein (VCP) gene.3, 4 Mutations in VCP have subsequently also been identified in patients with familial and sporadic ALS. Recently, two novel genes for MSP have been identified. Mutations in the prion-like domain (PrLD) of the RNA binding genes heterogeneous nuclear ribonucleoproteins A1 and A2B1 (hnRNPA1 and hnRNPA2B1) were shown to be associated with MSP pedigrees negative for VCP.5 Interestingly, several genes that have previously been implicated in ALS and FTD (TDP-43, FUS, TAF15, ESWR1)6 are also RNA binding genes with a PrLD (in which the majority of pathogenic mutations are found). Considering the clinical, pathological and genetic overlap between MSP and the classical forms of ALS, FTD and IBM, we decided to sequence the PrLD of hnRNPA1 and hnRNPA2B1 for mutations in a large population of sporadic and familial ALS, FTD and IBM patients from the Netherlands.


109

Mutations in hnRNPA1/A2B1 in ALS, FTD and IBM patients

METHODS

DNA samples of ALS and IBM patients were collected as part of the Prospective ALS study The Netherlands (PAN), a population-based case-control study including both patients with ALS and ALS mimics.7 Additionally, DNA samples were collected from FTD patients ascertained at the Erasmus Medical Center Rotterdam. Genetic analysis of the PrLD of hnRNPA1 (residues 251-320) and hnRNPA2B1 (residues 253-333) was performed by Sanger sequencing. Fifty-two of the 68 familial FTD patients were screened by next generation exome sequencing (for complete description of genetic analysis, see supplementary data).

RESULTS

A total of 135 familial ALS, 1084 sporadic ALS, 68 familial FTD, 74 sporadic FTD and 31 sporadic IBM patients from The Netherlands were included in this study (Table 8.1). No non-synonymous mutations were identified in the coding exons of the PrLD of hnRNPA1 and hnRNPA2B1. One potentially interesting splice variant was detected in a single case of familial FTD (hnRNPA2B1: c.695A>G) and is predicted as disease causing by mutationtaster.

Table 8.1 Baseline characteristics

Nu

mbe

r of

p

atie

nts

Fem

ale,

%

Age

at

incl

usi

on,

y, m

edia

n (r

ange

)

Age

at

onse

t, y

, m

edia

n (r

ange

)

Surv

ival

, y,

med

ian

(ran

ge)

Fam

ily

his

tory

of

dem

enti

a, %

Fam

ily

his

tory

of

AL

S, %

Familial ALS 135 45.2 63.9 (34-87) 59.7 (24-82) 3.2 (0.7-19) 30.6 100.0

Sporadic ALS 1084 39.9 65.2 (24-91) 63.7 (23-90) 2.6 (0.4-33) 26.9 0.0

Familial FTD 68 48.5 63.5 (38-77) 59.5 (36-73) 8.1 (2-24) 100.0 5.9

Sporadic FTD 74 59.6 62.4 (37-79) 57.8 (30-75) 10.1 (2-22) 0.0 1.0

IBM a 31 29.4 69.9 (52-86) - - - -aPatients were included at their first visit to the clinic. No information was available on age at onset and survival of the IBM patients.


CHAPTER 8

110

DISCUSSION

In this study we screened the largest cohort of ALS patients (n=1219) to date for mutations in hnRNPA1 and hnRNPA2B1. To our knowledge, this the first study in which considerable cohorts of FTD (n=144) and IBM patients (n=31) have been analyzed. We did not identify any non-synonymous mutations in our cohorts, which suggests that mutations in these genes are not a common cause of ALS, FTD and IBM in The Netherlands. In the initial publication, one mutation was found in 212 familial ALS cases and one mutation in 305 sporadic ALS cases, however segregation of hnRNP with ALS was not shown.5 A recent French study found no mutations in 17 patients with an MSP phenotype and 60 FTD and FTD-ALS patients.4 A subsequent Italian study did not find mutations in 221 familial ALS patients and 622 sporadic ALS patients.8 Combined over these three studies a total of 568 familial ALS patients and 2011 sporadic ALS patients have been screened. The overall impression of these data is that the frequency of hnRNPA1 and hnRNPA2B1 mutations in ALS is low (familial ALS: 1/568= 0.17% and sporadic ALS: 1/2011= 0.05%). Alternatively, the lack of replication in this study could theoretically be explained if the initial report were to be false positive, or more likely due to considerable differences in the frequency of these mutations across populations as has been shown previously for mutations in other ALS genes.9 It will therefore be interesting to see the results of mutational screening in other populations and larger cohorts of FTD and IBM cases. For now, it seems that in clinical practice screening for mutations in hnRNPA1 and hnRNPA2B1 should perhaps be reserved for those patients with a family history suggesting MSP.

Disclosure statementThe authors report no conflict of interest. All participants gave written informed consent and the Medical Ethics Review Boards of the participating institutions approved this study.


111


REFERENCES

1. Neumann M, et al. Ubiquitinated TDP-43 in frontotemporal lobar degeneration and amyotrophic lateral sclerosis. Science. 2006; 314: 130-3.

2. Weihl CC, et al. TDP-43 accumulation in inclusion body myopathy muscle suggests a common pathogenic mechanism with frontotemporal dementia. J Neurol Neurosurg Psychiatry. 2008; 79: 1186-9.

3. Watts GD, et al. Inclusion body myopathy associated with Paget disease of bone and frontotemporal dementia is caused by mutant valosin-containing protein. Nat Genet. 2004; 36: 377-81.

4. Le Ber I, et al. hnRNPA2B1 and hnRNPA1 mutations are rare in patients with "multisystem proteinopathy" and frontotemporal lobar degeneration phenotypes. Neurobiol Aging. 2014; 35: 934 e5-6.


6. King OD, et al. The tip of the iceberg: RNA-binding proteins with prion-like domains in neurodegenerative disease. Brain Res. 2012; 1462: 61-80.


8. Calini D, et al. Analysis of hnRNPA1, A2/B1, and A3 genes in patients with amyotrophic lateral sclerosis. Neurobiol Aging. 2013; 34: 2695 e11-2.

9. van Es MA, et al. Large-scale SOD1 mutation screening provides evidence for genetic heterogeneity in amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry. 2010; 81: 562-6


CHAPTER 8

112

SUPPLEMENTARY DATA

Genetic analysis

Sanger sequencingGenomic DNA was extracted from whole blood using standard procedures according to the chemagic DNA Blood5k Kit (magnetic bead DNA isolation procedure). DNA was whole genome amplified using a Qiagen Repli-g Mini kit according to manufacturer’s instructions using 50 ng of input genomic DNA per reaction. Polymerase chain reaction (PCR) was performed of the exonic sequences encoding the core PrLD and at least 50 base pairs of flanking intronic sequences in two amplicons for each of the two hnRNPs (Ensembl transcript ID ENSG00000135486 for hnRNPA1 and ENSG00000122566 for hnRNPA2B1). Sequences are listed in Supplementary Table S8.1. PCR was performed using a thermocycling program (96 ºC for 5 minutes; 35 cycles of 96 ºC for 30 seconds, 57 ºC for 45 seconds, 72 ºC for 1 minute; 72 ºC for 5 minutes and infinite hold at 8 ºC). PCR reactions consisted of 1.0 μL amplified DNA, 1.0 μL 10xNH4 reaction buffer, 0.2 μL dinucleotide triphosphate (dNTP; 10mM each), 0.4 μL dimethyl sulfoxide (DMSO), 0.3 μL MgCl2 (50mM), 0.1 μL Biotaq (5U/μL), 6.6 μL milli-Q, 0.2 μL of each primer (10 μM) in a total volume of 10 μL.PCR products were checked on a 1.2% agarose gel. BigDye Terminator 3.1 sequencing kit (Applied Biosystems, Foster City, CA, USA), DNA Analyzer 3730XL, and PolyPhred software version 6.18 (Washington University, Seattle, WA, USA, Nickerson et al., 1997) were used for sequencing and data analysis.

Whole exome sequencingWhole exome capture and sequencing were performed by Human Genomics Facility at Rotterdam. Exomes were captured by Nimblegen seqcap EZ human exome v2, and were sequenced with 100 base pair reads on the Illumina HiSeq2000 platform, according to the manufacturer’s protocol. Reads were mapped to the human reference genome sequence (assembly GRCh37/hg19) using the Burrows-Wheeler Alignment Tool.1 Base quality recalibration, local sequence realignment and variant filtering to minimize base calling and mapping errors were performed by Samtools2, Picard (http://picard.sourceforge.net)

Supplementary Table S8.1 Amplicons used to sequence exon 8 and 9 of hnRNPA1 and exon 9 and 10 of hnRNPA2B1 gene

Gene Exon Forward Reverse

hnRNPA1 8 ‘5 - GACCTTAGGCGCTTAGTTGATG ‘5 - GCCCAGACATAGCAGTTAAAGG

9 ‘5 - CCTCTTTACCACCTCCCTTG ‘5- TGCACTGCTCAGCTACATTAGG

hnRNPA2B1 9 ‘5 - AGCTGGAAATGGATGTGAGG ‘5 - ACCAAGGACTTAGGACAAAGC

10 ‘5 - CACCTGCAACCTTTATGTGG ‘5 - GCACTGCCCACAGTACAAAC


113


and Genome analysis Tool Kit (GATK)3 . The identified variants per individual were called by using GATK and annotated by ANNOVAR4. Variants with quality score < 50, quality over depth < 5.0, Strandbias > 0.75 and depth < 5.0 were filtered out. Variants in the hnRNPA1 and hnRNPA2B1 were examined on their frequency in dbSNP (http://www.ncbi.nlm.nih.gov/projects/SNP/), the 1000 genome project (www.1000genomes.org/) and the National Heart Lung Blood Institute Exome Variant Server (EVS) (https://evs.gs.washington.edu/EVS/). The predicted functional effects of the variants were assessed by Polyphen-2 (http://genetics.bwh.harvard.edu/pph2/), Sorting Intolerant from Tolerant (SIFT) (http://sift.jcvi.org/www/SIFT_enst_submit.html), PROVEAN (http://provean.jcvi.org/seq_submit.php) and Mutation Taster (www.mutationtaster.org).

REFERENCES

1. Li H, et al. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009; 25, 1754-1760.

2. Li H, et al. 1000 Genome Project Data Processing Subgroup. The Sequence Alignment/Map format and SAMtools. Bioinformatics. 2009; 25: 2078-2079.

3. McKenna A, et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 2010; 20: 1297-1303.

4. Wang K, et al. ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res. 2010; 38: e164.



Meinie Seelen1, Perry T.C. van Doormaal1, Wouter van Rheenen1, Reinoud J.P. Bothof1, Tim van Riessen1, William J. Brands1, Anneke J. van der Kooi2, Marianne de Visser2,

Nicol C. Voermans3, Jan H. Veldink1*, Leonard H. van den Berg1*, Michael A. van Es1*


2 Department of Neurology, Academic Medical Center, University of Amsterdam, The Netherlands

3 Department of Neurology, Donders Institute for Brain, Cognition and Behavior, Radboud University Nijmegen Medical Center, The Netherlands


In preparation

Large scale genetic screening in sporadic

ALS identifies modifiers in C9orf72 repeat carriers

CHAPTER 9


CHAPTER 9

116

ABSTRACT

Sporadic ALS is considered to be a complex disease with multiple genetic risk factors contributing to the pathogenesis. Identification of genetic risk factors that co-occur frequently could provide relevant insight into underlying mechanisms of motor neuron degeneration. To dissect the genetic architecture of sporadic ALS we undertook a large sequencing study in 755 apparently sporadic ALS cases and 959 controls, analyzing ten ALS genes: SOD1, C9orf72, TARDBP, FUS, ANG, CHMP2B, ATXN2, NIPA1, SMN1 and UNC13A. We observed sporadic cases with multiple genetic risk variants in 4.1% compared to 1.3% in controls. This difference was not in excess of what is to be expected by chance (binomial test, P = 0.59). We did observe a higher frequency than expected of C9orf72 repeat carriers with co-occurring susceptibility variants (ATXN2, NIPA1, SMN1; P = 0.001), which is mainly due to the co-occurrence of NIPA1 repeats in 15% of C9orf72 repeat carriers (P = 0.006).

INTRODUCTION

Amyotrophic lateral sclerosis (ALS) is a fatal neurological disorder characterized by motor neuron degeneration in the primary motor cortex, brainstem and spinal cord. ALS patients develop progressive weakness and spasticity eventually resulting in respiratory failure and death. Survival is approximately 3 years from symptom onset.1-3 To date, there is only one drug, riluzole, that can moderately slow disease progression.4

In about 5-10% of patients the disease is familial and the mode of transmission is mostly autosomal dominant.5 In the majority of cases there is no apparent family history and these cases are considered sporadic. Although the distinction between familial and sporadic seems straightforward, there is no clear definition of familial ALS and there is poor consensus amongst experts.6 Many familial ALS pedigrees demonstrate incomplete penetrance. Genetic modeling studies have shown that considering non-penetrance and small family size familial cases may well present as “apparently” sporadic.7 Even further complicating the matter is that there have been reports of ALS pedigrees with mutations in more than one ALS gene, suggesting that oligogenic inheritance also occurs.8-13 Therefore, it has been argued that the distinction between familial and sporadic ALS is rather arbitrary.7 In fact it has been proposed that this distinction might be artificial in all diseases. In an influential paper, Manolio et al. propose a model in which they present genetic risk factors on a sliding scale with on one extreme rare mutations (<1%) with very large effect (perhaps directly pathogenic) involved in Mendelian disorders and on the other extreme common polymorphisms (>5%) with small effect (odds ratios < 1.5) likely to be involved in complex / sporadic diseases.14 They argue that the majority of genetic risk factors are likely to be variants with intermediate frequency (1-5%) and larger effect (odds ratios >1.5). This model seems to translate well to ALS genetics.


117

Genetic modifiers of C9orf72 repeat carriers

Over the last few years great progress has been made in ALS genetics. There are over 20 familial ALS genes and several risk factors for sporadic ALS have now been identified by GWAS and candidate gene studies.15, 16 In an attempt to dissect the genetic architecture of sporadic ALS we undertook a large sequencing study in which we analyzed 10 ALS genes: high risk genes (rare mutations with large effect): SOD1, C9orf72, TARDBP, FUS, susceptibility genes with intermediate effect: ANG, CHMP2B, ATXN2, NIPA1, SMN1, and polymorphisms with small effect: a risk SNP in UNC13A. We hypothesized that we would find: 1) a low percentage of sporadic cases with mutations in high risk genes (familial cases presenting as sporadic) and 2) a significant number of sporadic cases with variants in multiple susceptibility genes and/or mutations in high risk genes combined with susceptibility genes (perhaps contributing to non-penetrance and phenotypic variability). The identification of genetic risk factors that co-occur frequently could provide relevant insight into the underlying mechanisms of motor neuron degeneration, which formed the rationale for this study.

METHODS

SubjectsSporadic ALS patients and control subjects were recruited as part of the Prospective ALS study The Netherlands, a population-based case-control study.1 All ALS patients included in this study fulfilled the revised El Escorial criteria for definite or probable ALS.38 ALS mimics and subjects with a known family history of ALS were excluded. Genomic DNA was extracted from whole blood by means of salting-out or magnetic beads procedures. All material was obtained with the ethical approval of the institutional review board of the University Medical Center Utrecht. All subjects provided written informed consent.

Gene selection To date, over 100 genes have been implicated in the cause of ALS.39 The level of supporting evidence for each gene or gene variant varies from small to overwhelming, and is in some cases contradictory.40 Therefore most authors assign ALS genes to different categories of certainty, although there is no consistent nomenclature. In this study, we divided genetic variants into three categories: 1) High risk variants, 2) Susceptibility variants and 3) Risk SNPs. Variants were grouped according to consensus in literature (directly causal / Mendelian or less certain (risk factor / putative ALS gene))15, 16, 41 minor allele frequency (MAF) and effect size. We considered rare variants (MAF <0.1% in controls) with large effect (causal / Mendelian) as high risk variants. Variants with low MAF (0.1 - 5.0% in controls) with intermediate effect (OR between 1.5 and 10.0) as susceptibility variants. Variants with MAF > 5.0% in controls and OR < 1.5 were termed risk SNPs.In total 10 genes were selected for analysis in this study. They were categorized as follows: 1) High risk genes: SOD1, TARDBP, FUS, C9orf72; 2) Susceptibility variants: ANG, CHMP2B, NIPA1, SMN1, ATXN2; and 3) Risk SNP: UNC13A. These genes were selected based on


CHAPTER 9

118

the following criteria: 1) the presence of variation in the gene in the Dutch population in previous studies. For instance, UBQLN2 mutations are a well-established, but rare cause of ALS and mutations in this gene are not found in The Netherlands.42 Therefore UBQLN2 was not analysed in this study. For this reason, genes such as OPTN, VCP, hnRNPA1 and hnRNPA2B1 were also not included in the study. And 2) the genes needed to be implicated in ALS by multiple or very large studies.

Genetic analysesTo complete the entire set of genes for each subject, we used previously obtained data and subsequently performed additional genetic analyses. Detailed information of the genetic analyses can be found in the Supplementary Material (Methods).In short, subjects were screened for mutations in SOD1 (NM_000454, exons 1-5), FUS (NM_004960, exons 5, 6, 14, 15), TARDBP (NM_007375, exon 6), ANG (NM_001145, exon 2) and CHMP2B (NM_014043, exons 1-6) by means of sanger sequencing as described previously.19, 43-46 These exons were selected since the vast majority of known pathogenic variants lie within these regions. Additional screening of FUS, TARDBP, ANG and CHMP2B was performed by multiplexed targeted resequencing, carried out on a MiSeq high-throughput next-generation sequencing platform (Illumina). Bar-coded paired-end sequencing libraries were prepared using a Truseq Custom Amplicon kit (Illumina). Sequence Analysis Viewer software (Illumina) was used to monitor the quality of the sequencing runs. Sequencing reads were mapped to the human genome reference build GRCh37 using Burrows-Wheeler Aligner (BWA v0.6.1). Subsequent depth of coverage, quality filters, variant calling and variant annotation were performed using SAMtools v0.1.19, GATK v3.2 and the 1000 Genomes project. The impact of each mutation on the structure, and function of the protein was predicted with PolyPhen-2 (PolyPhen-2 v2.2.2; http://genetics.bwh.harvard.edu/pph2/) and PMut (http://mmb2.pcb.ub.es:8080/PMut/).We performed fragment-length analyses of repeats in C9orf72 (NM_018325, long repeat = (GGGGCC) ≥30), ATXN2 (NM_002973, intermediate repeat = (CAG) ≥29) and NIPA1 (NM_144599, long repeat = (GCG) >8), as described previously.47-49 Copy number variations in SMN1 (NM_000344, i.e. >2 copies) were determined using multiplexed ligation-dependent probe amplification, as described previously.50 Lastly, a previously ALS-associated SNP in UNC13A (rs12608932, recessive model CC) was determined by use of a TaqMan allelic discrimination assay.30

Statistical analysisBinomial tests were performed to assess whether the observed frequency of co-occurring variants was in excess of what would be expected on the basis of chance. To calculate the expected frequency of co-occurring variants, we used the following formula: (the observed number of patients carrying a variant / the total number of patients) * (the observed number of controls carrying a variant / the total number of controls). This formula was used in order to take into account the higher frequency of just one variant in ALS patients (= frequency of variants in patients), multiplied by the chance probability of a second


119


variant (= frequency of variants in controls). To perform the binomial test, we then used the following formula in R: pbinom([observed number of patients with multiple variants], [total number of patients], [expected frequency], lower.tail = FALSE, log.p = FALSE). If the binomial p value is smaller than 0.05 this means that the observed frequency of co-occurring variants is higher than expected based on chance. We examined phenotypic associations for the combination of C9orf72 repeat expansions with susceptibility variants for age at onset, site of onset and survival. Statistical analysis was performed on SPSS software version 21 and R v3.0.2 (CRAN: http://www.r-project.org/).

RESULTS

A total of 755 sporadic ALS patients and 959 control subjects from The Netherlands were included in this study. Baseline characteristics are shown in Table 9.1. An overview of the genotyping results by gene is provided in Table 9.2. Approximately 7% of sporadic cases were found to carry variants in high risk ALS genes (SOD1, TARDBP, FUS and C9orf72). Repeat expansions in C9orf72 were the most common (6.1%). We identified a non-synonymous mutation in FUS in a single control. The frequency of variants in high risk ALS genes in controls was 0.1%. This difference between patients and controls was significant with P = 2.20 x 10-16. Variants in susceptibility genes ANG, CHMP2B, SMN1 duplications and repeat expansions in ATXN2 and NIPA1, were found in 14.9% of patients and in 8.4% of controls (P = 2.85 x 10-5). Lastly, 120 (15.9%) patients and 104 (10.8%) controls were homozygous for the UNC13A risk SNP (P = 0.002 (recessive model)).

Co-occurring ALS gene variantsIn 31 (4.1%) patients and in 13 (1.4%) controls we identified more than one variant in ALS-associated genes (Table 9.3). The higher frequency of ALS patients with multiple variants is statistically significant when applying a simple Fisher exact test with P = 7.39 x 10-4. When we subsequently performed a binomial test, in order to control for co-occurrence of multiple variants by chance (and taking the higher rate of having one ALS mutation in patients into account), this difference did not remain significant with P = 0.59.

Table 9.1 Baseline characteristics of study population

Sporadic ALS patients Control subjects

Subjects, No. 755 959

Female gender (%) 334 (44) 455 (47)

Age (y), median (IQR) 61 (53-69) 63 (56-70)

Bulbar site of onset, n (%) 254 (33.6)

Survival (m), median (IQR) 31 (21-45)

IQR, interquartile range. Age is depicted in years (y), survival in months (m).


CHAPTER 9

120

Table 9.2 Variants found by large scale genetic screening in sporadic ALS patients and control subjects

Gene Variant Exon ALS patients Controls Previous reports

SOD1 I99V 4 1 0 ALS43

E132K 5 1 0 Novel

Total (%) 2/755 (0.3) 0/959 (0.0)

FUS S115N 5 1 0 ALS46

S142N 5 0 1 Novel

R495X 14 1 0 ALS46, 51

Total (%) 2/753 (0.3) 1/943 (0.1)

TARDBP N352S 6 1 0 ALS46, 52

I383V 6 1 0 ALS45, 53

Total (%) 2/753 (0.3) 0/959 (0.0)

C9orf72 Long repeat 46 0 ALS/FTD20, 47, 54

Total (%) 46/755 (6.1) 0/959 (0.0)

ANG K17I 2 4 1 ALS/PD/CON19, 55

I46V 2 0 1 ALS/PD/CON19, 55

T80S 2 1 0 ALS19

F100I 2 1 0 ALS19

Total (%) 6/707 (0.9) 2/948 (0.2)

CHMP2B R22Q 2 1 0 ALS46

S103C 3 1 0 Novel

S194L 6 0 1 CON46

E201Q 6 1 0 Novel

Total (%) 3/738 (0.4) 1/928 (0.1)

ATXN2 Intermediate repeat 12 7 ALS48, 56

Total (%) 12/755 (1.6) 7/951 (0.7)

NIPA1 Long repeat 41 37 ALS49

Total (%) 41/740 (5.5) 37/956 (3.9)

SMN1 Duplications 50 33 ALS50, 57

Total (%) 50/755 (6.6) 33/959 (3.4)

UNC13A rs12608932 (CC) 120 104 ALS58

Total (%) 120/754 (15.9) 104/958 (10.8)

ALS, amyotrophic lateral sclerosis; FTD, frontotemporal dementia; PD, Parkinson’s disease; CON, controls. SOD1 (NM_000454, exons 1-5), FUS (NM_004960, exons 5, 6, 14, 15), TARDBP (NM_007375, exon 6), C9orf72 (NM_018325, long repeat = (GGGGCC) ≥30), ANG (NM_001145, exon 2), CHMP2B (NM_014043, exons 1-6), ATXN2 (NM_002973, intermediate repeat = (CAG) ≥29) and NIPA1 (NM_144599, long repeat = (GCG) >8), SMN1 (NM_000344, >2 copies), UNC13A (NM_001080421, homozygous SNP).


121


When we looked at subgroups, no co-occurring high risk variants (SOD1, FUS, TARDBP, C9orf72) were found in sporadic ALS patients (Table 9.3a). The frequency of high risk variants combined with susceptibility variants was significantly higher than would be expected on the basis of chance (P = 0.001, Table 9.3b), which is probably due to combinations with C9orf72 repeat expansions (11 out of 12 co-occurring variants). Interestingly, we observed a significantly lower rate of controls (P = 0.009) with variants in multiple high risk and susceptibility genes (without UNC13A SNP, expected 7 (0.7%) versus actually observed 1 (0.1%)). Clinical characteristics (i.e. gender, age at onset, site of onset, survival) of patients with multiple genetics variants are described in the Supplementary Material (Table S9.1).

C9orf72 repeat carriersConsidering the most prominent finding was the co-occurrence of C9orf72 repeat expansions with susceptibility variants (ATXN2, NIPA1, SMN1) and the UNC13A SNP, we performed additional analyses. The difference was mainly explained by the combination of C9orf72 and NIPA1 repeat expansions (binomial test, P = 5.73 x 10-4, Table 9.4). No significant difference in observed versus expected frequency was found for C9orf72 in combination with ATXN2, SMN1 or UNC13A (P = 0.06, P = 0.08, P = 0.13). Considering the relatively high frequency of C9orf72 and NIPA1 repeat expansions (15% of C9orf72 cases), we questioned whether the initial association of NIPA1 with ALS might be driven by a high frequency of coincidental C9orf72 repeat expansions and that the association with NIPA1 could be false positive. Therefore the association analysis on the NIPA1 data set was repeated after excluding C9orf72 positive cases and the association still remained, suggesting that there is a significant co-occurrence of 2 independently ALS-associated repeat expansions.There were two cases with a C9orf72 repeat expansion combined with variants in two other genes (C9orf72, ATXN2 and UNC13A; and C9orf72, SMN1 and UNC13A). For both cases statistical analysis suggested that these combinations were not likely to be a chance finding with P = 6.60 x 10-4 and P = 0.01. Phenotypic characteristics of all C9orf72 repeat carriers with concurrent possible genetic modifiers are shown in Table S9.2 (i.e. age at onset, site of onset, survival and co-morbid frontotemporal dementia). Unfortunately no detailed family history on cognitive impairment was available, since most patients were included many years ago. C9orf72 repeat carriers with a NIPA1 repeat expansion had an earlier mean age at onset (52 vs 60 years, P = 0.03) and more often a spinal onset (86 vs 53%, P = 0.21) compared to C9orf72 repeat carriers without a NIPA1 repeat expansion. C9orf72 repeat carriers with a concomitant UNC13A SNP more often had a bulbar onset (86 vs 33%, P = 0.01) and a shorter survival (26.3 vs 33.3 months, P = 0.48).


CHAPTER 9

122

Tab

le 9

.3 T

he o

bser

ved

and

expe

cted

co-

occu

rrin

g va

rian

ts in

spo

radi

c A

LS

pati

ents

and

con

trol

sub

ject

s

Var

ian

t 1

Var

ian

t 2

Var

ian

t 3

AL

S p

atie

nts

, n

(%)

Ex

pec

ted

freq

uen

cy

AL

S, n

(%)

a

Bin

omia

l p

-val

ueb

Co

ntr

ols

, n

(%)

A) C

ombi

nat

ion

of

hig

h r

isk

var

ian

ts

- -

-0

(0.0

)0

(0.0

3)-

0 (0

.0)

B) C

ombi

nat

ion

of

a h

igh

ris

k v

aria

nt

wit

h a

su

scep

tibi

lity

var

ian

t

SOD

1 (I

99V

)SM

N1

(dup

licat

ions

)1

0

C9o

rf72

(lon

g re

peat

)N

IPA

1 (lo

ng r

epea

t)7

0

C9o

rf72

(lon

g re

peat

)SM

N1

(dup

licat

ions

)3

0

C9o

rf72

(lon

g re

peat

)AT

XN

2 (in

term

edia

te r

epea

t)1

0

Tota

l12

(1.6

)4

(0.6

)0.

001

0 (0

.0)

C) C

ombi

nat

ion

of

any

hig

h r

isk

an

d su

scep

tibi

lity

var

ian

ts

SOD

1 (I

99V

)SM

N1

(dup

licat

ions

)1

0

AN

G (K

17I)

ATX

N2

(inte

rmed

iate

rep

eat)

SMN

1 (d

uplic

atio

ns)

10

C9o

rf72

(lon

g re

peat

)N

IPA

1 (lo

ng r

epea

t)7

0

C9o

rf72

(lon

g re

peat

)SM

N1

(dup

licat

ions

)3

0

C9o

rf72

(lon

g re

peat

)AT

XN

2 (in

term

edia

te r

epea

t)1

0

ATX

N2

(inte

rmed

iate

rep

eat)

SMN

1 (d

uplic

atio

ns)

10

ATX

N2

(inte

rmed

iate

rep

eat)

NIP

A1

(long

rep

eat)

01

Tota

l14

(1.9

)12

(1.7

)0.

271

(0.1

)

D) C

ombi

nat

ion

of

hig

h r

isk

var

ian

ts, s

usc

epti

bili

ty v

aria

nts

an

d ri

sk S

NP

SOD

1 (I

99V

)SM

N1

(dup

licat

ions

)1

0

TAR

DB

P (N

352S

)U

NC

13A

(rs1

2608

932)

10


123


AN

G (K

17I)

ATX

N2

(inte

rmed

iate

rep

eat)

SMN

1 (d

uplic

atio

ns)

10

CH

MP2

B (E

201Q

)U

NC

13A

(rs1

2608

932)

10

C9o

rf72

(lon

g re

peat

)N

IPA

1 (lo

ng r

epea

t)7

0

C9o

rf72

(lon

g re

peat

)SM

N1

(dup

licat

ions

)U

NC

13A

(rs1

2608

932)

10

C9o

rf72

(lon

g re

peat

)SM

N1

(dup

licat

ions

)2

0

C9o

rf72

(lon

g re

peat

)AT

XN

2 (in

term

edia

te r

epea

t)U

NC

13A

(rs1

2608

932)

10

C9o

rf72

(lon

g re

peat

)U

NC

13A

(rs1

2608

932)

50

ATX

N2

(inte

rmed

iate

rep

eat)

SMN

1 (d

uplic

atio

ns)

10

ATX

N2

(inte

rmed

iate

rep

eat)

NIP

A1

(long

rep

eat)

01

ATX

N2

(inte

rmed

iate

rep

eat)

UN

C13

A (r

s126

0893

2)0

2

NIP

A1

(long

rep

eat)

UN

C13

A (r

s126

0893

2)4

6

SMN

1 (d

uplic

atio

ns)

UN

C13

A (r

s126

0893

2)6

4

Tota

l31

(4.1

)32

(4.2

)0.

5913

(1.4

)

Four

dif

fere

nt g

roup

s of

co-

occu

rrin

g va

rian

ts a

re s

how

n, a

ddin

g ne

w p

ossi

ble

com

bina

tion

s of

ris

k va

rian

ts in

eac

h st

ep in

a s

lidin

g sc

ale

of c

erta

inty

. a To

cal

cula

te th

e ex

pect

ed fr

eque

ncy

of c

o-oc

curr

ing

vari

ants

, we

used

the

follo

win

g fo

rmul

a: (t

he o

bser

ved

num

ber

of p

atie

nts

carr

ying

one

var

iant

/

the

tota

l num

ber

of p

atie

nts)

* (t

he o

bser

ved

num

ber

of c

ontr

ols

carr

ying

one

var

iant

/ th

e to

tal n

umbe

r of

con

trol

s). T

his

form

ula

was

use

d in

ord

er

to t

ake

into

acc

ount

the

hig

her

freq

uenc

y of

just

one

var

iant

in A

LS

pati

ents

(= f

requ

ency

of

vari

ants

in p

atie

nts)

tim

es t

he c

hanc

e pr

obab

ility

of

a se

cond

var

iant

(= fr

eque

ncy

of v

aria

nts

in c

ontr

ols)

. b A

bin

omia

l tes

t was

per

form

ed to

com

pare

the

obse

rved

freq

uenc

y of

co-

occu

rrin

g va

rian

ts in

sp

orad

ic A

LS

pati

ents

wit

h th

e ca

lcul

ated

exp

ecte

d fr

eque

ncy.

A

LS,

am

yotr

ophi

c la

tera

l scl

eros

is. S

OD

1 (N

M_0

0045

4, e

xons

1-5

), C

9orf

72 (N

M_0

1832

5, lo

ng r

epea

t = (G

GG

GC

C) ≥

30),

AN

G (N

M_0

0114

5, e

xon

2), C

HM

P2B

(NM

_014

043,

exo

ns 1

-6),

AT

XN

2 (N

M_0

0297

3, in

term

edia

te r

epea

t = (C

AG

) ≥29

) and

NIP

A1

(NM

_144

599,

long

rep

eat =

(GC

G) >

8),

SMN

1 (N

M_0

0034

4, >

2 co

pies

), U

NC

13A

(NM

_001

0804

21, h

omoz

ygou

s SN

P).


CHAPTER 9

124

DISCUSSION

In this study we attempted to dissect the genetics of sporadic ALS by analyzing 10 ALS genes in a large cohort of population based cases and controls. We did not find an overall increased risk of co-occurring variants in sporadic ALS patients. But we do present compelling statistical evidence for an excess of concomitant mutations in C9orf72 repeat carriers.Approximately 7% of our sporadic cases were found to carry a variant in a high risk ALS gene. We did not observe sporadic cases with simultaneous mutations in more than one high risk ALS gene. Previous studies in familial ALS have shown pedigrees with variants in multiple high risk ALS genes suggesting that in a percentage of familial ALS there is oligogenic inheritance.8-10, 17, 18 The fact that these double mutations do occur in familial ALS but not in sporadic ALS may suggest that the co-occurrence of mutations in 2 high risk genes results in a familial rather than a sporadic presentation.Sporadic ALS is considered to be a complex disease with multiple genetic risk factors contributing to the disease. We therefore expected to find sporadic cases with multiple genetic risk variants. Indeed, we observed double mutations in 4.1% of patients compared to 1.3% of controls. Despite the 3-fold higher frequency in cases, this difference was not statistically significant when we applied a binomial test, taking the higher rate of having one ALS mutation in patients into account. With this test we compared the frequency of observed double mutations in cases to the expected number of double mutations in cases (using the frequency of mutations in controls). Although correcting for chance co-occurrence is necessary, our approach may be too strict considering that we also correct for combinations that have never been observed, such as homozygous NIPA1 expansions. Likewise, in a recent meta-analysis on ANG mutations in ALS and Parkinson’s disease data from over 15,000 individuals was available in which no homozygous ANG mutations were observed.19 Hence, we may be overcorrecting for merely theoretical possibilities. This seems to be reinforced by the fact that the observed frequency of double mutations was significantly lower than was to be expected in controls. We therefore hypothesize that perhaps certain combinations of mutations may be more relevant than we can

Table 9.4 Frequency of C9orf72 repeat expansions with co-occurring variants (ATXN2, NIPA1, SMN1, UNC13A)

Genes sporadic ALS (n=755) (%)

C9orf72 carriers (n=46) (%)

Expected frequency (%)

Binomial P valuea

C9orf72 + ATXN2 1 (0.13) 1 (2.2) 0.4 (0.05) 0.06

C9orf72 + NIPA1 7 (0.93) 7 (15.2) 1.8 (0.24) 5.73 x 10-4

C9orf72 + SMN1 3 (0.40) 3 (6.5) 1.6 (0.21) 0.08

C9orf72 + UNC13A 7 (0.93) 7 (15.2) 5.0 (0.66) 0.13

ALS, amyotrophic lateral sclerosis. aP values result from binomial test, comparing the observed frequency in sporadic ALS versus the expected frequency (calculated for all sporadic ALS patients).


125


statistically demonstrate. The most straightforward approach to test this hypothesis is to increase sample size. In this study we present compelling statistical evidence for an excess of concomitant mutations in combination with C9orf72 repeat expansions. Although C9orf72 was initially discovered in ALS-FTD, the gene has now been implicated in many different neurodegenerative and psychiatric diseases including Alzheimer’s disease, Parkinsonism, Huntington’s disease phenocopies, schizophrenia and bipolar disorder.20-23 This very large phenotypic variability associated with C9orf72 repeat expansions is poorly understood. One of the hypotheses is that additional genetic variants determine phenotype in C9orf72 carriers. Indeed there are multiple case reports of C9orf72 cases with additional mutations in other ALS or FTD associated genes (i.e. ANG, TARDBP, FUS, SOD1, VAPB, OPTN, UBQLN2, MAPT, GRN, DAO).8-11, 13, 17, 24-26 However, only a few studies have systematically analyzed multiple genes in cohorts. A French study on the role of ATXN2 expansions in neurodegeneration found C9orf72 expansions in 5.5% of sporadic ALS cases with ATXN2 expansions, and ATXN2 expansions in 1.8% of C9orf72 positive sporadic ALS cases, with even higher frequencies in familial ALS and FTD cases.27 Another recent study by van Blitterswijk et al. also provided evidence for ATXN2 as a disease modifier of C9orf72.28 In our data set we did not observe a significant co-occurrence of ATXN2 in C9orf72 repeat carriers (n = 1). However, the frequency of co-occurring ATXN2 and C9orf72 repeat expansions in our study is comparable to that of the other studies. Therefore our findings do not contradict the previous studies. Several studies demonstrated that UNC13A homozygous ALS cases have a shorter survival.29, 30 A recent study demonstrated that multiple genetic factors influence phenotypic features in C9orf72 ALS.31 UNC13A was found to negatively influence survival in C9orf72-ALS. In this study we also observed a shorter survival for C9orf72 – UNC13A ALS cases (26.3 vs 33.3 months).In the current study, we observed a high frequency of NIPA1 repeat expansions in C9orf72 positive sporadic ALS cases (15.2% compared to 5.5% in all other sporadic ALS cases and only 3.9% in controls). This difference was significant and represents the discovery of a novel phenotypic modifier of the C9orf72 phenotype. The NIPA1 expansion is highly interesting as it is solely made up of alanines, the majority being encoded by a polymorphic (GCG)n repeat (most frequently (GCG)7 and (GCG)8).

32 A shared feature of all polyalanine disease pathogenesis is prominent protein aggregation.33 In vitro experiments have shown that peptides containing 7-15 alanine repeats undergo variable levels of conformational transition from a monomeric α-helix to a predominant β-sheet. However, when the repeat size is >15, there is complete conversion from monomer to β-sheet.34 In vivo experiments show that expanded polyalanine tracts have a pronounced tendency to adopt β-sheet complexes that promote strong protein–protein interactions, leading to insoluble protein assemblies. The level of protein aggregation in polyalanine diseases correlates with the size of the polyalanine expansion tract.33 There is also considerable evidence suggesting that misfolded polyalanine-containing proteins are targeted for degradation by the ubiquitin-proteasome system.35


CHAPTER 9

126

Allelic mutations in NIPA1 cause Spastic Paraplegia 6 (SPG6), which is an upper motor neuron syndrome.36 Copy number variations in NIPA1 are associated with ALS.37 And our data shows that NIPA1 repeat expansions are a risk factor for ALS independent of C9orf72. There is thus considerable evidence that variation in NIPA1 affects the motor system in various ways. It therefore seems plausible that a NIPA1 repeat expansion in the context of a C9orf72 repeat expansion would drive towards a motor neuron disease phenotype. Indeed, it seems there is a relatively consistent phenotype in C9orf72 – NIPA1 ALS cases (relatively young, spinal onset).To our knowledge, this is the first report on ALS cases carrying 3 different ALS risk factors. Two out of these three cases are C9orf72 positive, which again suggests that it are the additional genetic factors that determine phenotype in C9orf72 carriers.

In summary, there is an increasing number of reports on both familial and sporadic ALS patients with mutations in multiple ALS (risk) genes. To date, very few studies have systematically addressed this phenomenon, but it seems that collaborative and larger studies will be able to identify frequently co-occurring variants, which in turn could provide novel inroads for creating disease models and perhaps therapeutic strategies. Although the high phenotypic variability associated with C9orf72 repeat expansions is not well understood, there is mounting evidence that additional genetic factors determine phenotype. Several genes have been implicated to date, including UNC13A and ATXN2 repeat expansions. In this study we identified a novel modifier, NIPA1.


127


REFERENCES


2. Pugliatti M, et al. Amyotrophic lateral sclerosis in Sardinia, insular Italy, 1995-2009. J Neurol. 2013; 260: 572-9.

3. Rooney J, et al. Survival analysis of irish amyotrophic lateral sclerosis patients diagnosed from 1995-2010. PLoS One. 2013; 8: e74733.

4. Miller RG, et al. Riluzole for amyotrophic lateral sclerosis (ALS)/motor neuron disease (MND). Cochrane Database Syst Rev. 2012; 3: CD001447.

5. Byrne S, et al. Rate of familial amyotrophic lateral sclerosis: a systematic review and meta-analysis. J Neurol Neurosurg Psychiatry. 2011; 82: 623-7.

6. Byrne S, et al. Absence of consensus in diagnostic criteria for familial neurodegenerative diseases. J Neurol Neurosurg Psychiatry. 2012; 83: 365-7.

7. Al-Chalabi A and Lewis CM. Modelling the effects of penetrance and family size on rates of sporadic and familial disease. Hum Hered. 2011; 71: 281-8.

8. Chio A, et al. ALS/FTD phenotype in two Sardinian families carrying both C9orf72 and TARDBP mutations. J Neurol Neurosurg Psychiatry. 2012; 83: 730-3.


10. Millecamps S, et al. Phenotype difference between ALS patients with expanded repeats in C9orf72 and patients with mutations in other ALS-related genes. J Med Genet. 2012; 49: 258-63.

11. King A, et al. Mixed tau, TDP-43 and p62 pathology in FTLD associated with a C9orf72 repeat expansion and p.Ala239Thr MAPT (tau) variant. Acta Neuropathol. 2013; 125: 303-10.

12. Testi S, et al. Co-Occurrence of the C9orf72 Expansion and a Novel GRN Mutation in a Family with Alternative Expression of Frontotemporal Dementia and Amyotrophic Lateral Sclerosis. J Alzheimers Dis. 2014.

13. van Blitterswijk M, et al. VAPB and C9orf72 mutations in 1 familial amyotrophic lateral sclerosis patient. Neurobiol Aging. 2012; 33: 2950 e1-4.

14. Manolio TA, et al. Finding the missing heritability of complex diseases. Nature. 2009; 461: 747-53.

15. Renton AE, et al. State of play in amyotrophic lateral sclerosis genetics. Nat Neurosci. 2014; 17: 17-23.

16. Leblond CS, et al. Dissection of genetic factors associated with amyotrophic lateral sclerosis. Exp Neurol. 2014.

17. Kenna KP, et al. Delineating the genetic heterogeneity of ALS using targeted high-throughput sequencing. J Med Genet. 2013; 50: 776-83.

18. CadyJ,etal.ALSonsetisinfluencedbytheburdenofrarevariantsinknownALSgenes.Ann Neurol. 2014.

19. van Es MA, et al. Angiogenin variants in Parkinson disease and amyotrophic lateral sclerosis. Ann Neurol. 2011; 70: 964-73.





24. Lattante S, et al. Contribution of major amyotrophic lateral sclerosis genes to the etiology of sporadic disease. Neurology. 2012; 79: 66-72.


CHAPTER 9

128

25. Kaivorinne AL, et al. Novel TARDBP sequence variant and C9orf72 repeat expansion in a family with frontotemporal dementia. Alzheimer Dis Assoc Disord. 2014; 28: 190-3.

26. van Blitterswijk M, et al. C9orf72 repeat expansions in cases with previously identifiedpathogenic mutations. Neurology. 2013; 81: 1332-41.

27. Lattante S, et al. Contribution of ATXN2 intermediary polyQ expansions in a spectrum of neurodegenerative disorders. Neurology. 2014; 83: 990-5.

28. vanBlitterswijkM,etal.Ataxin-2aspotentialdiseasemodifierinC9orf72expansioncarriers.Neurobiol Aging. 2014; 35: 2421 e13-7.

29. ChioA,etal.UNC13AinfluencessurvivalinItalianamyotrophiclateralsclerosispatients:apopulation-based study. Neurobiol Aging. 2013; 34: 357 e1-5.

30. DiekstraFP,etal.UNC13Aisamodifierofsurvivalinamyotrophiclateralsclerosis.Neurobiol Aging. 2012; 33: 630 e3-8.

31. vanBlitterswijkM,etal.GeneticmodifiersincarriersofrepeatexpansionsintheC9orf72gene. Mol Neurodegener. 2014; 9: 38.

32. Chai JH, et al. Identification of four highly conserved genes between breakpoint hotspotsBP1 and BP2 of the Prader-Willi/Angelman syndromes deletion region that have undergone evolutionarytranspositionmediatedbyflankingduplicons.Am J Hum Genet. 2003; 73: 898-925.

33. Messaed C and Rouleau GA. Molecular mechanisms underlying polyalanine diseases. Neurobiol Dis. 2009; 34: 397-405.

34. Shinchuk LM, et al. Poly-(L-alanine) expansions form core beta-sheets that nucleate amyloid assembly. Proteins. 2005; 61: 579-89.

35. Abu-Baker A, et al. Involvement of the ubiquitin-proteasome pathway and molecular chaperones in oculopharyngeal muscular dystrophy. Hum Mol Genet. 2003; 12: 2609-23.

36. Rainier S, et al. NIPA1 gene mutations cause autosomal dominant hereditary spastic paraplegia (SPG6). Am J Hum Genet. 2003; 73: 967-71.

37. Blauw HM, et al. A large genome scan for rare CNVs in amyotrophic lateral sclerosis. Hum Mol Genet. 2010; 19: 4091-9.


39. Lill CM, et al. Keeping up with genetic discoveries in amyotrophic lateral sclerosis: the ALSoD and ALSGene databases. Amyotroph Lateral Scler. 2011; 12: 238-49.

40. Abel O, et al. Credibility analysis of putative disease-causing genes using bioinformatics. PLoS One. 2013; 8: e64899.

41. Ajroud-Driss S and Siddique T. Sporadic and hereditary amyotrophic lateral sclerosis (ALS). Biochim Biophys Acta. 2014.

42. van Doormaal PT, et al. UBQLN2 in familial amyotrophic lateral sclerosis in The Netherlands. Neurobiol Aging. 2012; 33: 2233 e7- e8.

43. van Es MA, et al. Large-scale SOD1 mutation screening provides evidence for genetic heterogeneity in amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry. 2010; 81: 562-6.

44. Groen EJ, et al. FUS mutations in familial amyotrophic lateral sclerosis in the Netherlands. Arch Neurol. 2010; 67: 224-30.

45. Ticozzi N, et al. Mutational analysis of TARDBP in neurodegenerative diseases. Neurobiol Aging. 2011; 32: 2096-9.

46. van Blitterswijk M, et al. Genetic overlap between apparently sporadic motor neuron diseases. PLoS One. 2012; 7: e48983.

47. van Rheenen W, et al. Hexanucleotide repeat expansions in C9orf72 in the spectrum of motor neuron diseases. Neurology. 2012; 79: 878-82.

48. VanDammeP, et al. ExpandedATXN2CAG repeat size inALS identifies genetic overlapbetween ALS and SCA2. Neurology. 2011; 76: 2066-72.

49. Blauw HM, et al. NIPA1 polyalanine repeat expansions are associated with amyotrophic lateral sclerosis. Hum Mol Genet. 2012; 21: 2497-502.

50. Blauw HM, et al. SMN1 gene duplications are associated with sporadic ALS. Neurology. 2012; 78: 776-80.

51. Waibel S, et al. Novel missense and truncating mutations in FUS/TLS in familial ALS. Neurology. 2010; 75: 815-7.


129


52. Kuhnlein P, et al. Two German kindreds with familial amyotrophic lateral sclerosis due to TARDBP mutations. Arch Neurol. 2008; 65: 1185-9.

53. Rutherford NJ, et al. Novel mutations in TARDBP (TDP-43) in patients with familial amyotrophic lateral sclerosis. PLoS Genet. 2008; 4: e1000193.

54. Renton AE, et al. A hexanucleotide repeat expansion in C9orf72 is the cause of chromosome 9p21-linked ALS-FTD. Neuron. 2011; 72: 257-68.

55. Greenway MJ, et al. ANG mutations segregate with familial and 'sporadic' amyotrophic lateral sclerosis. Nat Genet. 2006; 38: 411-3.

56. Elden AC, et al. Ataxin-2 intermediate-length polyglutamine expansions are associated with increased risk for ALS. Nature. 2010; 466: 1069-75.

57. Corcia P, et al. Abnormal SMN1 gene copy number is a susceptibility factor for amyotrophic lateral sclerosis. Ann Neurol. 2002; 51: 243-6.

58. vanEsMA,etal.Genome-wideassociationstudyidentifies19p13.3(UNC13A)and9p21.2assusceptibility loci for sporadic amyotrophic lateral sclerosis. Nat Genet. 2009; 41: 1083-7.


CHAPTER 9

130

SUPPLEMENTARY MATERIAL

Supplementary Methods – Genetic analysesSanger SequencingVenous blood samples were drawn using 10-mL EDTA tubes, and genomic DNA was extracted from whole blood using a standard salting-out procedure or by using magnetic beads (chemagic kit, Perkin Elmer). All five exons of SOD1 (NM_000454) were sequenced using BigDye Terminator 3.1 technology, after initial touchdown PCR amplification, as described previously.1 The following primers were used: SOD1-1-F, CGTCGTAGTCTCCTGCAGCG, and SOD1-1-R, GCGGGCGACCCGCTCCTAGC; SOD1-2-F, GGGTAAAGGTAAATCAGCTG, and SOD1-2-R, ATCTAACTAGGGTGAACAAG; SOD1-3-F, CCCAGAAGTCGTGATGCAGG, and SOD1-3-R, CCATATGAACTCCAGAAAGC; SOD1-4-F, TGCAAATTTGTGTCTACTCAGTC, and SOD1-4-R, CCGCGACTAACAATCAAAGTC; SOD1-5-F, GGTAGTGATTACTTGACAGC, and SOD1-5-R, CAGGTACTTTAAAGCAACTC. PCR products were sequenced on an ABI3730xl sequencer (Applied Biosystems). Each mutation was confirmed on genomic DNA. Sequencing was performed on FUS (NM_004960, exons 5, 6, 14, 15), using a 96-capillary DNA Analyzer 3730XL and a BigDye Terminator 3.1 sequencing kit (Applied Biosystems, Foster City, California) as described previously.2 The following primers were used in this study for exon 5, 6, 14 and 15 respectively: FUS-5-F, CACGACGTTGTAAAACGACTGGACTCCACTAAAAGTGAAAGG, and FUS-5-R, GGATAACAATTTCACACAGGAAAATGGGCTGCAGACAAAG; FUS-6-F, GAGGGTTCCTGTCTTGTTTC, and FUS-6-R, CCTCACAGATCCCTAGACAAC; FUS-14-F, CACGACGTTGTAAAACGACGAGCTGGGACCAAAGAATCC, and FUS-14-R, GGATAACAATTTCACACAGGCCCCTGAGTTAATTTTCCTTCC; FUS-15-F, CACGACGTTGTAAAACGACGGTAGGAGGGGCAGATAGG, and FUS-15-R, GGATAACAATTTCACACAGGCTTGGGTGATCAGGAATT. All mutations were confirmed in independent experiments on genomic DNA. Sequence data were analyzed in PolyPhred.Mutational screening of exon 6 of TARDBP (NM_007375) was performed by touchdown PCR using the following primers TDP43-6-F, AGTAAAACGACGGCCAGTTGAATCAGTGGTTTAATCTTCTTTG; and TDP43-6-R, GCAGGAAACAGCTATGACCAAAATTTGAATTCCCACCATTC as described previously.3 These primers anneal to adjacent intronic and 3’UTR regions of exon 6 and contain 5’ tails encoding M13 forward and reverse. PCR-products were subsequently purified by incubation with Exonuclease I and Shrimp Alkaline Phosphatase, sequenced with M13 primers using the BigDyeTerminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA) and then resolved by capillary electrophoresis on an ABI 3730XL DNA Analyzer (Applied Biosystem). Sequence analysis was performed using the PHRED/PHRAP/Consed software suite (http://www.phrap.org/) and


131


variations in the sequences were identified with the Polyphred v6.15 software.Sequencing was performed on the single coding exon of ANG (NM_001097577, exon 2), using a 96- capillary DNA Analyzer 3730XL (Applied Biosystems, Foster City, CA) and BigDye Terminator 3.1 chemistry as described previously.4 The following primers were used in this study: ANG-1-F, GTTCTTGG GTCTACCACACC and ANG-1-R, AATGGAAGGCAAGGA CAGC. The sequences were aligned using the PHRED/PHRAP/Consed package (http://www.phrap.org/), and variants were identified using the software application PolyPhred. When a variant was identified, this was confirmed by independent experiments using newly prepared samples from stock DNA.Coding regions of all six exons CHMP2B (NM_014043.3) were screened for mutations using touchdown PCR, as described previously.5 The following primers were used: CHMP2B-1-F, CCGCAGACGTGAGGAAAG, and CHMP2B-1-R, CTCCAGGGACAGTAGGCAGA; CHMP2B-2-F, GCGCCCAGCCAATATAAGAT, and CHMP2B-2-R, GCCATGTGCCTTCTTCCTAGT; CHMP2B-3-F, CTTCATGATCGGGGACAAAG, and CHMP2B-3-R, CAGGAGGTGCTTTTAAATCTGC; CHMP2B-4-F, TTTGATGTGTTCCCTTTTGACTT, and CHMP2B-4-R, TCATCATTTCTGCCTTCGTG; CHMP2B-5-F, TTCACTGAGTTTGCCTTCTGT, and CHMP2B-5-R, CGTGCATTAGGAAACATTTGG; CHMP2B-6-F, GGAGGTGCATGGTTTTTATTTC, and CHMP2B-6-R, TTGGCAGCTGTAACCACCTA (for PCR), GAAATCTGCACTGTGCTTGG (for sequencing). Sanger sequencing and data analysis were performed with BigDye Terminator 3.1 sequencing kit (Applied Biosystems, Foster City, California), DNA Analyzer 3730XL (Applied Biosystems) and PolyPhred. Each mutation was confirmed on genomic DNA.

High-Throughput Next-Generation SequencingAdditional screening for non-synonymous mutations in FUS (exon 5, 6, 14, 15), TARDBP (exon 6), ANG (exon 2) and CHMP2B (exon 1-6) was carried out on a MiSeq high-throughput next-generation sequencing platform (Illumina). Designstudio (Illumina) was used to create a Truseq Custom Amplicon project, which involved the use of specific DNA oligo stretches to sequence all exons and flanking region of the abovementioned genes of interest. Standard Truseq Custom Amplicon Library Preparation protocol was followed to create genetic libraries containing all regions of interest of our samples. By barcoding each sample with unique indexed primers, ligated to the target regions and amplified by means of standard PCR, we multiplexed 95 samples of patients and control subjects per sequencing run. We chose a paired-end read with a 2x 250 basepair read length for the approximately 28 amplicons (amplicon length 425 bp) representing our regions of interest, which ensured excellent coverage of over 500 resequencing reactions for each amplicon. Sequence Analysis Viewer software (Illumina) was used to monitor the quality of every sequencing run, and secondary analyses were performed on MiSeq Reporter software to accomplish variant calling for every sample. Sequencing reads were mapped to the human


CHAPTER 9

132

genome reference build GRCh37 using Burrows-Wheeler Aligner (BWA v0.6.1). Subsequent depth of coverage, quality filters, variant calling and variant annotation were performed using SAMtools v0.1.19, GATK v3.2 and the 1000 Genomes project. Using SAMtools the average per sample depth of coverage was calculated to be 4332. GATK was used to perform local indel realignment, recalibrate base quality scores, call variants and assign sample genotypes. Variants would “pass” quality filters based on whether they met a series of quality control criteria. These criteria included a minimum variant quality score of 30, and adapted GATK filters (“QD” <2.0, “FS” >500, “MQ” <40.0, “HaplotypeScore” >300, “MQRankSum” <-12.5 and “ReadPosRankSum” <-8.0, values). Next, those variants not characterised as silent, intronic or non-coding were added to the final dataset.

Fragment-length Analysis A repeat primed PCR was performed to assess the GGGGCC repeat in C9orf72 (NM_018325), as described elsewhere.6 In short, a primed PCR protocol was used on 50ng genomic DNA with the following primer sequences: forward primer 5’ – 6FAM-AGTCGCTAGAGGCGAAAGC – 3’, reverse primer 5’ – TACGCATCCCAGTTTGAGACGGGGGCCGGGGCCGGGGCCGGGG – 3’, and anchor primer 5’ – TACGCATCCCAGTTTGAGACG – 3’. Fragment analysis was accomplished on an ABI3730xl sequencer and fragment sizes were analysed with GeneMapper software version 3.7. Furthermore, all samples were genotyped at the repeat sites at least two times. Alleles with 30 or more GGGGCC hexanucleotide repeats were defined as expanded. The CAG repeat of ATXN2 (NM_002973) was amplified using following primers: forward primer 5’ – 6FAM-GGGCCCCTCACCATGTCG – 3’ and reverse primer 5’ – CGGGCTTGCGGACATTGG – 3’, as described previously.7 The GCG repeat in exon 1 of NIPA1 (NM_144599) was genotyped with the following primer sequences: a fluorescently labelled forward primer 5’ – 6FAM-GCCCCTCTTCCTGCTCCT – 3’ and reverse primer with sequence 5’ – CGATGCCCTTCTTCTGTAGC – 3’, as described previously.8 Repeat lengths were determined on a ABI3130xl sequencer. For determination of the fragment length Peak Scanner software version 1.0 (Applied Biosystems) was used.

Multiplexed Ligation-dependent Probe AmplificationCopy number variation in SMN1 (NM_000344) was identified by multiplexed ligation-dependent probe amplification (MLPA) assays were run using standard protocols (www.mlpa.com), as described previously.9 We used the SALSA P060 MLPA kit (MRC Holland, the Netherlands), containing 2 probes specifically targeted to SMN1, and control probes targeted to other chromosomal loci for normalization and assay quality control. A total of 50 –100 ng of genomic DNA was used in each MLPA assay. Data normalization and analysis were performed with GeneMarker software (SoftGenetics, State College, PA) using standard parameters.


133


TaqMan Allelic Discrimination AssayTaqMan allelic discrimination assay was used to determine the single nucleotide polymorphism (SNP) in UNC13A at position rs12608932, as previously described.10 The following primer and probe sequences were used [VIC/FAM]: ATCCATCCACCCATCAATTTATCCA[A/C]CCATCCATTTTTCGTCTGTCCACCA. Allelic PCR products were analysed using specific probes on the ABI Prism 7900HT Sequence Detection System. SDS software version 2.3 (Applied Biosystems) was used to analyze allelic variant calls for each sample.

REFERENCES

1. van Es MA, Dahlberg C, Birve A, Veldink JH, van den Berg LH, Andersen PM. Large-scale SOD1 mutation screening provides evidence for genetic heterogeneity in amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry 2010;81:562-566.

2. Groen EJ, van Es MA, van Vught PW, et al. FUS mutations in familial amyotrophic lateral sclerosis in the Netherlands. Archives of neurology 2010;67:224-230.

3. Ticozzi N, LeClerc AL, van Blitterswijk M, et al. Mutational analysis of TARDBP in neurodegenerative diseases. Neurobiology of aging 2011;32:2096-2099.

4. van Es MA, Schelhaas HJ, van Vught PW, et al. Angiogenin variants in Parkinson disease and amyotrophic lateral sclerosis. Annals of neurology 2011;70:964-973.

5. van Blitterswijk M, Vlam L, van Es MA, et al. Genetic overlap between apparently sporadic motor neuron diseases. PloS one 2012;7:e48983.

6. van Rheenen W, van Blitterswijk M, Huisman MH, et al. Hexanucleotide repeat expansions in C9ORF72 in the spectrum of motor neuron diseases. Neurology 2012;79:878-882.

7. Van Damme P, Veldink JH, van Blitterswijk M, et al. Expanded ATXN2 CAG repeat size in ALSidentifiesgeneticoverlapbetweenALSandSCA2.Neurology2011;76:2066-2072.

8. Blauw HM, van Rheenen W, Koppers M, et al. NIPA1 polyalanine repeat expansions are associated with amyotrophic lateral sclerosis. Human molecular genetics 2012;21:2497-2502.

9. Blauw HM, Barnes CP, van Vught PW, et al. SMN1 gene duplications are associated with sporadic ALS. Neurology 2012;78:776-780.

10. vanEsMA,VeldinkJH,SarisCG,etal.Genome-wideassociationstudy identifies19p13.3(UNC13A) and 9p21.2 as susceptibility loci for sporadic amyotrophic lateral sclerosis. Nature genetics 2009;41:1083-1087.


CHAPTER 9

134

Table S9.1 Clinical information on ALS patients with co-occurring variants

Gene 1 Gene 2 Gene 3 Gender Age at Onset (y)

Site of Onset Survival (m)

SOD1 (I99V) SMN1 M 60 Bulbar 51

TARDBP (N352S) UNC13A M 42 Cervical 58

ANG (K17I) ATXN2 SMN1 F 77 Bulbar 21

CHMP2B (E201Q) UNC13A M 75 Spinal 4

C9orf72 ATXN2 UNC13A M 62 Bulbar 22

C9orf72 NIPA1 F 50 Lumbosacral 32

C9orf72 NIPA1 M 52 Bulbar 22


C9orf72 NIPA1 M 57 Lumbosacral 30

C9orf72 NIPA1 M 52 Lumbosacral 14


C9orf72 NIPA1 M 37 Cervical 87

C9orf72 SMN1 UNC13A F 63 Bulbar 30

C9orf72 SMN1 M 63 Lumbosacral 55

C9orf72 SMN1 F 66 Lumbosacral 61

C9orf72 UNC13A M 54 Bulbar 26

C9orf72 UNC13A F 56 Bulbar 23

C9orf72 UNC13A M 60 Bulbar 33

C9orf72 UNC13A M 53 Lumbosacral 14

C9orf72 UNC13A F 56 Bulbar 36

ATXN2 SMN1 F 51 Cervical 67

NIPA1 UNC13A M 50 Cervical 48

NIPA1 UNC13A F 77 Lumbosacral 13

NIPA1 UNC13A M 74 Bulbar 16

NIPA1 UNC13A M 60 Bulbar 30

SMN1 UNC13A M 55 Lumbosacral 34

SMN1 UNC13A F 62 Bulbar 19

SMN1 UNC13A M 70 Cervical 29

SMN1 UNC13A F 71 Cervical 45

SMN1 UNC13A F 69 Thoracic 37

SMN1 UNC13A F 73 Thoracic 14

M, male; F, female; y, years; m, months


135


Table S9.2 Possible genetic modifiers of C9orf72 repeat carriers

No Age at onset, mean (SD)

Bulbar site of onset, n (%)

Survival, m, mean (SD)

Co-morbid FTD, n (%)

NIPA1 - 39 60.3 (7.9) 18 (47) 31.0 (12.7) 4 (11)

NIPA1 + 7 52.7 (8.2) 1 (14) 38.0 (30.9) 1 (14)

P value* 0.03 0.21 0.95 0.59

ATXN2 - 45 59.3 (8.4) 18 (40) 32.4 (16.5) 5 (11)

ATXN2 + 1 62.0 (-) 1 (100) 22.0 (-) 0 (0)

P value* 0.52 0.41 0.44 0.89

SMN1 - 43 59.1 (8.5) 18 (42) 31.7 (16.4) 4 (9)

SMN1 + 3 64.0 (1.7) 1 (33) 42.5 (17.7) 1 (33)

P value* 0.1 1 0.28 0.30

UNC13A - 39 59.7 (8.9) 13 (33) 33.3 (17.4) 5 (13)

UNC13A + 7 57.7 (3.9) 6 (86) 26.3 (7.5) 0 (0)

P value* 0.58 0.01 0.48 0.42

FTD, frontotemporal dementia; m, months; patients with (NIPA1+) and without (NIPA1-) a NIPA1 long repeat; patients with (ATNX2+) and without (ATXN2-) intermediate repeat; patients with (SMN1+) and without (SMN1-) SMN1 duplications; patients with (UNC13A+) and without (UNC13A-) the recessive model of the UNC13A SNP (rs12608932).*P value of age at onset amd survival is based on Mann Withney U test, P value of site of onset and co-morbid FTD is based on Fisher exact test.



General discussion

CHAPTER 10


CHAPTER 10

138

In this thesis I describe the results of studies aiming to identify risk factors for amyotrophic lateral sclerosis (ALS). We have used a population-based case-control design to perform (1) epidemiological risk factor studies, examining lifestyle factors and environmental exposures, and (2) genetic studies determining genetic risk variants in ALS. Within these studies we have established novel exposures, which had not been implicated in ALS before, including exposure to traffic related air pollution and diesel motor exhaust. We were able to identify these exposures in two independent populations, in The Netherlands and Ireland, as well as through different study approaches (residential and occupational exposure assessment), presenting convincing evidence for the positive association with ALS risk. These and other ALS associated risk factors described in this thesis (e.g. physical activity, head trauma, high dietary fat intake, low BMI and cholesterol levels, and low alcohol consumption) have provided us with new clues for pathophysiological mechanisms, such as mitochondrial dysfunction, oxidative stress, neuroinflammation and glutamate excitotoxicity. Moreover, we have shown that ALS has a complex etiology in which multiple genetic (and most probably also lifestyle and environmental) factors co-occur to cause ALS. We specifically identified co-occurring repeat expansions in C9orf72 and NIPA1. Taking all studies together, we have generated new hypotheses for future functional biological studies to elucidate pathophysiological mechanisms leading to this complex and devastating disease.

Risk factors for amyotrophic lateral sclerosisPhysical activityPhysical activity as a risk factor for ALS is a frequently discussed topic, fuelled by anecdotal observations of famous athletes diagnosed with ALS, such as the 1930s American baseball player Lou Gehrig.1 Second to that, the association is biologically plausible, because of existing cellular and genetic evidence indicating that vigorous physical activity may induce oxidative stress and glutamate excitotoxicity.2 Also, several genes associated with the response to exercise, i.e ciliary neurotrophic factor, leukaemia inhibitory factor and vascular endothelial growth factor 2, have been identified as possible modifiers of ALS susceptibility.3-5

In this thesis, we report an increased risk of ALS with higher levels of leisure time physical activity (chapter 2). However, the lack of an association with occupational physical activity and the absence of a dose-response relationship strengthens the hypothesis that increased physical activity is not causally related with ALS risk, but that it may be a mutual genetic profile or lifestyle both promoting physical activity and increasing ALS susceptibility (“born to run” versus “run to death”). Moreover, multiple epidemiological studies were performed on the association between physical activity and ALS, which have shown conflicting results from harmful,6-8 null9-11 to beneficial effects,12 which emphasizes the complex relation between physical activity and ALS.


139

Discussion

Head traumaWe observed an increased risk of ALS after prior head trauma (chapter 3). This is consistent with several small studies and anecdotal reports suggesting that athletes exposed to repeated (minor) head trauma (e.g. in boxing, rugby, ice hockey, soccer players) were associated with long-term risks of neurodegenerative diseases.13, 14 For example, boxing has been known to cause chronic traumatic encephalopathy. In these patients with chronic traumatic encephalopathy, TDP-43 positive inclusions were found in the brain, a pathological hallmark of aggregated proteins that is frequently found in patients with ALS as well.15, 16 Moreover, a complex interplay of pathological mechanisms have been suggested following traumatic brain injury, such as glutamate excitotoxicity, oxidative stress, neuroinflammation and mitochrondrial dysfunction.17-19 An important issue to consider in the association between head trauma and ALS is reverse causality. The exact time of symptom onset of ALS is difficult to pinpoint, and incipient ALS may lead to for example tripping and subsequently head trauma. We therefore also analyzed the risk of ALS and head trauma occurring at least five years prior to symptom onset. In this additional analysis we still found an increased risk of ALS, making the issue of reverse causality less likely. Nevertheless, as was proposed in a recent letter to The Lancet Neurology, sound, evidence-based, high-quality studies are needed to investigate the long-term consequences of repeated, low-intensity head trauma in professional sportspeople and the associated long-term risks of developing devastating and life-threatening neurodegenerative disorders, such as ALS.13

Premorbid low cholesterol levels, low BMI and a high dietary fat intakeIn chapter 3, we observed a decreased risk of ALS in individuals with hypercholesterolemia or individuals using statins, indicating a relatively favorable lipid profile prior to onset in at least a subpopulation of ALS. This favorable lipid profile is consistent with the significantly lower premorbid body mass index (BMI) we found in ALS patients compared to controls. Previous studies support this finding with lower ALS rates among overweight and obese individuals and reports of ALS patients being more likely to have always been slim.8, 20

There is, however, also evidence for a high dietary fat and high caloric intake in ALS patients before onset of symptoms (chapter 4). This increased risk of ALS with high dietary fat intake has been reported previously in a smaller case-control study in the United States.21 The imbalance in intake (high fat/caloric diet) and output (low cholesterol levels/BMI) implies a role for an altered energy metabolism prior to symptom onset of ALS. It seems that patients with ALS “burn” more calories in a resting state as well as during exercise, also revered to as hypermetabolic.22 Furthermore, there is a growing body of evidence from animal studies in which mouse models of ALS show metabolic alterations with markedly increased resting and total energy expenditure, and increased lipolysis, premorbidly as well as after symptom onset.23, 24 It has been suggested that mitochondrial uncoupling protein 3 (UCP3) plays a role in this increased energy expenditure, since higher levels of expression of UCP3 have been found both in an animal model of ALS and in


CHAPTER 10

140

human biopsies.25-27 This implicates an important role of mitochondria in ALS pathogenesis. However, further clinical and experimental studies are needed to concentrate on the complex relations between (defective) energy metabolism and ALS. The hypothesis of an increased energy metabolism provides for new therapeutic options. It has previously been reported that mild obesity was associated with greater survival in patients with ALS,28 and that a high caloric diet increased survival in a mouse model.23 Compensating for the increased energy expenditure in ALS by initiating a high caloric diet may therefore prolong survival in ALS patients. Furthermore, a small randomized phase 2 clinical trial showed that there might be a more favorable outcome in patients on a hyper-carbohydrate hypercaloric diet compared to patients on a isocaloric diet or a hyper-fat hypercaloric diet.29 More research, however, is needed to specify which diet may be the most beneficial for patients with ALS.

AlcoholWe observed a decreased risk of ALS with a higher intake of alcohol (chapter 4). A finding that has not been reported in two previous epidemiological studies on alcohol consumption.30, 31 Still, functional biological studies did reveal a potential neuroprotective effect of the constituents of red wine. A lyophilized extract of red wine, which contains several antioxidant compounds, was able to block glutamate-induced apoptosis in cerebellar granule neurones.32 Furthermore, an in vivo experiment carried out on mutant SOD1 mice showed that survival in mice fed with lyophilized red wine was significantly increased compared to untreated animals. In our study, however, the association between intake of alcohol and the risk of ALS was independent of the intake of red wine, and so the association cannot be attributed only to the possible protective effect of antioxidants in red wine.

Smoking, diesel motor exhaust and air pollutionThus far, the only widely accepted environmental risk factor in ALS is smoking.33 Within the Prospective ALS study the Netherlands (PAN), we previously reported an increased risk of ALS with smoking.34 In this thesis, we observed an increased ALS risk in individuals exposed to diesel motor exhaust by occupation in two independent populations, in The Netherlands and in Ireland (chapter 5). Interestingly, in chapter 6 in a residential exposure study, we observed an increased ALS risk in individuals who are exposed to traffic related ambient air pollution, specifically PM2.5absorbance and nitrogen oxides. These observations parallel the observations with smoking. Smoking increases the risk of ALS through several potential mechanisms, including neuroinflammation, oxidative stress and direct neurotoxicity caused by fine particles, heavy metals and other chemical compounds present in cigarette smoke.35, 36 Ultrafine particles can circumvent the blood-brain barrier by deposition on the olfactory mucosa of the nasal region.37-39 These particles are then translocated along the olfactory nerve into the olfactory bulb of the brain, and may travel transneuronally to more distal sites within the brain.40, 41 Similar pathological mechanisms of neuroinflammation and oxidative stress, are suggested for diesel motor exhaust and


141

Discussion

traffic related air pollution.37, 39 Importantly, diesel motor exhaust and traffic related air pollution are environmental risk factors for ALS which are ubiquitous and can be modified at a population level. This adds to the necessity of regulatory public health interventions on air pollution exposure levels. It might be interesting to study whether patient groups with a specific gene mutation are at increased risk of developing ALS while exposed to smoking and/or diesel motor exhaust.

Electromagnetic fieldsStudying a different residential exposure, extremely low frequency electromagnetic fields (ELF-EMF), did not show an increased ALS risk (chapter 7), which is in concordance with previous residential exposure reports.42-44 However, occupational exposure studies primarily conclude that exposure to ELF-EMF is a risk factor for ALS.45-47 This discrepancy may be due to an increased risk of electric shocks or higher exposure levels of ELF-EMF in electrical occupations. Animal studies suggested that if ELF-EMF exposure would have an effect on developing ALS, it may cause damage to motor neurons through oxidative stress, though convincing evidence for this underlying biological mechanism is still lacking.48

hnRNPA1/A2B1Recently, new variants were identified in the prion-like domain of heterogeneous nuclear ribonucleoproteins (hnRNPs) A1 and A2B1 in families with multisystem proteinopathy (MSP).49 MSP is a clinical syndrome incorporating ALS, frontotemporal dementia, inclusion body myositis and Paget’s disease of the bone. In chapter 8 of this thesis we did not identify any mutations in hnRNPA1 and hnRNPA2B1 in ALS, FTD and IBM (non-MSP) patients in the Netherlands, also meaning that we did not find evidence for genetic pleiotropy of these genes. Few other studies assessed the frequency of mutations in hnRNPA1 and hnRNPA2B1 in ALS patients.49-52 Combined over these studies a total of 598 familial ALS patients and 2142 sporadic ALS patients have been screened, and only two patients carried a mutation (0.17% in familial ALS and 0.05% in sporadic ALS). The overall impression of these data is that the frequency of hnRNPA1 and hnRNPA2B1 mutations in ALS is low or population-specific. For now, it seems that in clinical practice screening for mutations in hnRNPA1 and hnRNPA2B1 should perhaps be reserved for those patients with a family history suggesting MSP.

Disease model in ALSOligogenic model In ALS patients with the same genetic background, the phenotypic heterogeneity can still be large, with for example on one end young onset and on the other end late onset or no disease onset at all. This heterogeneity is even expanding beyond ALS, showing pleiotropy. For example, in C9orf72 repeat carriers, this pleiotropy contains ALS and frontotemporal dementia, possibly even expanding to Parkinson’s disease, schizophrenia and bipolar disorder.53-56 The phenotype the C9orf72 repeat carriers express may be dependent on other co-occurring factors.


CHAPTER 10

142

Previous studies in familial ALS have shown pedigrees with variants in multiple high risk ALS genes suggesting that in a percentage of familial ALS there is oligogenic inheritance.57-61 In chapter 9, we studied the genetic architecture of sporadic ALS. Sporadic ALS is considered to be a complex disease with multiple genetic risk factors contributing to the disease. We observed co-occurring mutations in 4.1% of patients compared to 1.3% of controls. Despite the almost 3-fold higher frequency in cases, this difference was not in excess of what is to be expected by chance (binomial test, P = 0.59). We did, however, observe a higher frequency than expected of C9orf72 repeat carriers with co-occurring susceptibility variants ATXN2, NIPA1 and SMN1 (P = 0.001), which was mainly due to the co-occurrence of NIPA1 repeats in 15% of C9orf72 repeat carriers. This represents the discovery of a novel phenotypic modifier of the C9orf72 phenotype and emphasizes the theory of an oligogenic, or even polygenic, model in ALS.

Multistep modelA recent paper published in The Lancet Neurology assessed a hypothetically multistep model in ALS.62 This model has previously been applied to cancer epidemiology, and is about multiple lifestyle, environmental and genetic factors sequentially needed to cause the disease (e.g. a multistep process). In this paper, multiple ALS populations were assessed and a six step process was suggested to be needed to develop ALS.62 Interestingly, this multistep model was calculated based on age at onset and incidence data from several population based ALS studies, including the PAN, and is consistent with the oligogenic architecture described in my thesis. It would be interesting to identify the environmental and life style factors necessary on top of two or three mutations, which in different combinations might drive ALS causation. Clinical syndromeAs mentioned above, ALS is considered a complex disorder with large phenotypic diversity in between patients, e.g. in age at onset, site of onset, involvement of upper and lower motor neurons, cognitive impairment and survival. Furthermore there is heterogeneity in lifestyle, environmental and genetic risk factors, which may be explained by the quality (e.g. prevalent patient cohorts) and size of the studies performed. However, an alternative explanation for these heterogeneous results would be that ALS is not a single disease with a unique, yet unidentified cause, but a clinical syndrome in which multiple pathogenic pathways may lead to the phenotype of ALS. The combination of a multistep model and the concept of ALS as a clinical syndrome may explain why there is large phenotypic diversity in between patients and may explain the heterogenic results in risk factor studies.

Future directionsIdentifying subtypes of ALSAs ALS appears to be a complex and heterogeneous disease, it becomes increasingly important to identify subtypes of ALS by a more comprehensive approach. Subgroup


143

Discussion

analysis can be performed using latent class cluster analysis, identifying clusters of ALS patients with an identical cause. This analysis can be based on either lifestyle/environmental, genetic or phenotypic factors. The clusters will eventually be characterized on the strongest of these factors. This will lead to more homogeneous clusters or subtypes of ALS, which will improve power and provide for more solid and consistent results in future studies using these clusters. This approach is also compatible with the hypothesis of a multistep model in which variable combinations of factors lead to the development of ALS.

Gene-environment interaction studiesIn gene-environment interaction studies we aim to find an increased disease risk when both the genetic variant and the lifestyle/environmental factor are present. For example, a study in Parkinson’s disease assessed occupational exposures (such as pesticides, metals and solvents) and their interaction with variants in genes influencing metabolism of chemicals.63 They found possible interaction effects between GSTM1 genotype (a genetic risk factor for Parkinson’s disease) and solvents exposure. This approach of hypothesis driven gene-environment interactions would be interesting to apply to ALS research as well, for example mutations in SOD1, FUS, TARDBP or C9orf72 repeat expansion and their interaction with air pollution or head trauma. In which case, the disease risk would be higher when both factors are present compared to when no or just one of the factors is present. A second way to approach this analysis is to perform a genome-exposome wide interaction study. With the newest techniques and the availability of large amounts of lifestyle/environmental and genetic data, this hypothesis-free method would gain in associations otherwise not found in solely genome wide or exposome wide studies. Last, it would be good to combine the cluster analysis of ALS with genetic and lifestyle/environmental factors to identify the factors that lead to ALS subtypes.Ultimately, we aim to understand the biological pathways that lead to motor neuron degeneration by identifying as many risk factors for ALS as needed and by performing integrated gene-environment interaction analyses, if possible within identified subtypes. The final goal is to translate these findings to clinical uses and to provide for new treatment strategies or even preventive measures for this devastating disease.


CHAPTER 10

144

REFERENCES

1. Lewis M and Gordon PH. Lou Gehrig, rawhide, and 1938. Neurology. 2007; 68: 615-8.2. Harwood CA, et al. Physical activity as an exogenous risk factor in motor neuron disease

(MND): a review of the evidence. Amyotroph Lateral Scler. 2009; 10: 191-204.3. LambrechtsD,etal.VEGFisamodifierofamyotrophiclateralsclerosisinmiceandhumans

and protects motoneurons against ischemic death. Nat Genet. 2003; 34: 383-94.4. Al-ChalabiA,etal.Ciliaryneurotrophicfactorgenotypedoesnotinfluenceclinicalphenotype

in amyotrophic lateral sclerosis. Ann Neurol. 2003; 54: 130-4.5. Zheng C, et al. Vascular endothelial growth factor prolongs survival in a transgenic mouse

model of ALS. Ann Neurol. 2004; 56: 564-7.6. Beghi E, et al. Amyotrophic lateral sclerosis, physical exercise, trauma and sports: results of a

population-based pilot case-control study. Amyotroph Lateral Scler. 2010; 11: 289-92.7. Strickland D, et al. Physical activity, trauma, and ALS: a case-control study. Acta Neurol Scand.

1996; 94: 45-50.8. Scarmeas N, et al. Premorbid weight, body mass, and varsity athletics in ALS. Neurology. 2002;

59: 773-5.9. Longstreth WT, et al. Risk of amyotrophic lateral sclerosis and history of physical activity: a

population-based case-control study. Arch Neurol. 1998; 55: 201-6.10. Valenti M, et al. Amyotrophic lateral sclerosis and sports: a case-control study. Eur J Neurol.

2005; 12: 223-5.11. Veldink JH, et al. Physical activity and the association with sporadic ALS. Neurology. 2005; 64:

241-5.12. Pupillo E, et al. Physical activity and amyotrophic lateral sclerosis: a European population-

based case-control study. Ann Neurol. 2014; 75: 708-16.13. Pearce N, et al. Sports-related head trauma and neurodegenerative disease. Lancet Neurol.

2014; 13: 969-70.14. Gavett BE, et al. Chronic traumatic encephalopathy: a potential late effect of sport-related

concussive and subconcussive head trauma. Clin Sports Med. 2011; 30: 179-88, xi.15. McKee AC, et al. TDP-43 proteinopathy and motor neuron disease in chronic traumatic

encephalopathy. J Neuropathol Exp Neurol. 2010; 69: 918-29.16. Neumann M, et al. Ubiquitinated TDP-43 in frontotemporal lobar degeneration and

amyotrophic lateral sclerosis. Science. 2006; 314: 130-3.17. Mazzeo AT, et al. The role of mitochondrial transition pore, and its modulation, in traumatic

brain injury and delayed neurodegeneration after TBI. Exp Neurol. 2009; 218: 363-70.18. Lenzlinger PM, et al. Markers for cell-mediated immune response are elevated in cerebrospinal

fluidandserumafterseveretraumaticbraininjuryinhumans.J Neurotrauma. 2001; 18: 479-89.

19. Arundine M and Tymianski M. Molecular mechanisms of glutamate-dependent neurodegeneration in ischemia and traumatic brain injury. Cell Mol Life Sci. 2004; 61: 657-68.

20. O’Reilly EJ, et al. Premorbid body mass index and risk of amyotrophic lateral sclerosis. Amyotroph Lateral Scler Frontotemporal Degener. 2013; 14: 205-11.

21. Nelson LM, et al. Population-based case-control study of amyotrophic lateral sclerosis in western Washington State. II. Diet. Am J Epidemiol. 2000; 151: 164-73.

22. Dupuis L, et al. Energy metabolism in amyotrophic lateral sclerosis. Lancet Neurol. 2011; 10: 75-82.

23. Dupuis L, et al. Evidence for defective energy homeostasis in amyotrophic lateral sclerosis: benefitofahigh-energydietinatransgenicmousemodel.Proc Natl Acad Sci U S A. 2004; 101: 11159-64.

24. Dodge JC, et al. Metabolic signatures of amyotrophic lateral sclerosis reveal insights into disease pathogenesis. Proc Natl Acad Sci U S A. 2013; 110: 10812-7.

25. Li J, et al. Reducing systemic hypermetabolism by inducing hypothyroidism does not prolong survival in the SOD1-G93A mouse. Amyotroph Lateral Scler. 2012; 13: 372-7.


145

Discussion

26. Dupuis L, et al. Up-regulation of mitochondrial uncoupling protein 3 reveals an early muscular metabolic defect in amyotrophic lateral sclerosis. FASEB J. 2003; 17: 2091-3.

27. Clapham JC, et al. Mice overexpressing human uncoupling protein-3 in skeletal muscle are hyperphagic and lean. Nature. 2000; 406: 415-8.

28. Reich-Slotky R, et al. Body mass index (BMI) as predictor of ALSFRS-R score decline in ALS patients. Amyotroph Lateral Scler Frontotemporal Degener. 2013; 14: 212-6.

29. Wills AM, et al. Hypercaloric enteral nutrition in patients with amyotrophic lateral sclerosis: a randomised, double-blind, placebo-controlled phase 2 trial. Lancet. 2014; 383: 2065-72.

30. Nelson LM, et al. Population-based case-control study of amyotrophic lateral sclerosis in western Washington State. I. Cigarette smoking and alcohol consumption. Am J Epidemiol. 2000; 151: 156-63.

31. Okamoto K, et al. Lifestyle factors and risk of amyotrophic lateral sclerosis: a case-control study in Japan. Ann Epidemiol. 2009; 19: 359-64.

32. Esposito E, et al. Lyophilized red wine administration prolongs survival in an animal model of amyotrophic lateral sclerosis. Ann Neurol. 2000; 48: 686-7.


34. de Jong SW, et al. Smoking, alcohol consumption, and the risk of amyotrophic lateral sclerosis: a population-based study. Am J Epidemiol. 2012; 176: 233-9.


36. Alonso A, et al. Association of smoking with amyotrophic lateral sclerosis risk and survival in men and women: a prospective study. BMC Neurol. 2010; 10: 6.


38. OberdorsterG,etal.Translocationof inhaledultrafineparticles to thebrain. Inhal Toxicol. 2004; 16: 437-45.

39. TonelliLHandPostolacheTT.Airborneinflammatoryfactors:“fromthenosetothebrain”.Front Biosci (Schol Ed). 2010; 2: 135-52.

40. Tjalve H and Henriksson J. Uptake of metals in the brain via olfactory pathways. Neurotoxicology. 1999; 20: 181-95.

41. Elder A, et al. Translocation of inhaled ultrafinemanganese oxide particles to the centralnervous system. Environ Health Perspect. 2006; 114: 1172-8.


43. Huss A, et al. Residence near power lines and mortality from neurodegenerative diseases: longitudinal study of the Swiss population. Am J Epidemiol. 2009; 169: 167-75.

44. Marcilio I, et al. Adult mortality from leukemia, brain cancer, amyotrophic lateral sclerosis andmagneticfieldsfrompowerlines:acase-controlstudyinBrazil.Rev Bras Epidemiol. 2011; 14: 580-8.

45. HussA,etal.OccupationalexposuretomagneticfieldsandelectricshocksandriskofALS:The Swiss National Cohort. Amyotroph Lateral Scler Frontotemporal Degener. 2014: 1-6.

46. Davanipour Z, et al. Amyotrophic lateral sclerosis and occupational exposure to electromagnetic fields.Bioelectromagnetics. 1997; 18: 28-35.

47. Gunnarsson LG, et al. A case-control study of motor neurone disease: its relation to heritability, and occupational exposures, particularly to solvents. Br J Ind Med. 1992; 49: 791-8.

48. FaloneS,etal.Chronicexposureto50Hzmagneticfieldscausesasignificantweakeningofantioxidant defence systems in aged rat brain. Int J Biochem Cell Biol. 2008; 40: 2762-70.


50. Le Ber I, et al. hnRNPA2B1 and hnRNPA1 mutations are rare in patients with “multisystem proteinopathy” and frontotemporal lobar degeneration phenotypes. Neurobiol Aging. 2014; 35: 934 e5-6.

51. Calini D, et al. Analysis of hnRNPA1, A2/B1, and A3 genes in patients with amyotrophic lateral sclerosis. Neurobiol Aging. 2013; 34: 2695 e11-2.


CHAPTER 10

146

52. Soong BW, et al. Extensive molecular genetic survey of Taiwanese patients with amyotrophic lateral sclerosis. Neurobiol Aging. 2014; 35: 2423 e1-6.






58. Kenna KP, et al. Delineating the genetic heterogeneity of ALS using targeted high-throughput sequencing. J Med Genet. 2013; 50: 776-83.

59. Millecamps S, et al. Phenotype difference between ALS patients with expanded repeats in C9orf72 and patients with mutations in other ALS-related genes. J Med Genet. 2012; 49: 258-63.

60. Chio A, et al. ALS/FTD phenotype in two Sardinian families carrying both C9orf72 and TARDBP mutations. J Neurol Neurosurg Psychiatry. 2012; 83: 730-3.

61. CadyJ,etal.ALSonsetisinfluencedbytheburdenofrarevariantsinknownALSgenes.Ann Neurol. 2014.

62. Al-Chalabi A, et al. Analysis of amyotrophic lateral sclerosis as a multistep process: a population-based modelling study. Lancet Neurol. 2014; 13: 1108-13.

63. Dick FD, et al. Gene-environment interactions in parkinsonism and Parkinson’s disease: the Geoparkinson study. Occup Environ Med. 2007; 64: 673-80.


147



Nederlandse samenvatting - Summary in Dutch

Dankwoord - Acknowledgements

About the author

ADDENDUM


CHAPTER 10

150

NEDERLANDSE SAMENVATTING

Titel: Risicofactoren voor amyotrofische laterale sclerose

IntroductieAmyotrofische laterale sclerosis (ALS) is een progressieve neurodegeneratieve aandoening waarbij de motorische zenuwcellen in de hersenen, hersenstam en het ruggenmerg langzaam afsterven. De ziekte kenmerkt zich door toenemende spierzwakte in armen en benen, moeite met spreken en slikken en zwakte van de ademhalingsspieren. Het beloop van de ziekte is zeer variabel. Gemiddeld overlijden patiënten 3 jaar na ontstaan van de eerste verschijnselen, waarbij twintig procent van de patiënten langer leeft dan 5 jaar. Jaarlijks krijgen 400-500 Nederlanders de diagnose ALS. Naast de motorische uitval ontwikkelt een klein deel van de patiënten ook een stoornis in het denken of het gedrag, ook wel fronto-temporale dementie (FTD) genoemd, een vorm van dementie met gedragsveranderingen. Tot op heden is er geen curatieve behandeling voor ALS: het sinds 1995 geregistreerde medicijn Riluzole verlengt het leven gemiddeld slechts met drie maanden.Een klein percentage patiënten met ALS heeft een erfelijke vorm, waarbij er meerdere personen binnen één familie gedurende hun leven ALS ontwikkelen (familiaire ALS). Bij deze patiënten is er sprake van een belangrijke genetische factor (DNA-afwijking) als oorzaak van de ziekte, die wordt overgeërfd binnen de familie. Bij de meeste patiënten (in 90-95%) gaat het echter om de sporadische vorm, waarbij ALS verder niet binnen de familie voorkomt. De oorzaak van sporadische ALS is waarschijnlijk meer complex: een combinatie van risicofactoren die gezamenlijk de ziekte veroorzaken. Hierbij gaat het zowel om leefstijlfactoren, als omgevingsfactoren, als genetische factoren. In de drie verschillende delen van dit proefschrift worden deze risicofactoren uiteengezet: in deel I leefstijlfactoren, in deel II omgevingsfactoren, en in deel III genetische factoren.

Het onderzoek beschreven in dit proefschrift is onderdeel van de Prospectieve ALS studie Nederland (PAN studie). In deze studie worden sinds 2006 prospectief alle nieuw-gediagnosticeerde ALS patiënten in Nederland geïncludeerd. Via de huisarts worden controlepersonen benaderd die gematcht zijn aan de patiënt op basis van geslacht en leeftijd. Van zowel de patiënten als de controlepersonen wordt informatie over risicofactoren verzameld door gebruik te maken van uitgebreide en gestructureerde vragenlijsten en wordt bloed afgenomen voor DNA onderzoek.

Deel I - LeefstijlfactorenFysieke inspanning is één van de veelbesproken factoren in het internationale ALS onderzoek. De relatie met fysieke inspanning is eerder geopperd in een Italiaanse studie waarin een verhoogde incidentie van ALS werd beschreven onder professionele voetballers. Tevens zijn er diverse bekende sporters die gediagnosticeerd zijn met ALS, zoals Lou Gehrig, een Amerikaanse honkbalspeler naar wie de ziekte in Amerika vernoemd


151

Nederlandse samenvatting

is, of de recent gediagnosticeerde Nederlandse profvoetballer Fernando Ricksen. In hoofdstuk 2 van dit proefschrift hebben we onderzocht of er een relatie is tussen de mate van fysieke inspanning en het risico op ALS. Hieruit blijkt dat patiënten met ALS in hun vrije tijd (sport en hobby’s) meer fysieke inspanning leveren dan controlepersonen (gemiddelde cumulatieve MET score 1.51 vs. 1.25; MET = metabole equivalent van fysieke activiteiten). Patiënten leveren echter niet meer fysieke inspanning in hun beroep of ondergaan niet vaker extreme inspanningen, zoals het lopen van een marathon. Tevens hebben we geen dosis-respons relatie gevonden, dat wil zeggen dat het risico op ALS niet toeneemt naarmate de mate van fysieke inspanning toeneemt. Concluderend hebben we in ons onderzoek geen oorzakelijk verband kunnen aantonen tussen fysieke inspanning en ALS. Mogelijk is er wel een gezamenlijke basis (predispositie) die er enerzijds voor zorgt dat iemand aanleg heeft om fysiek actief te zijn, zoals in het sporten, en anderzijds een verhoogd risico op ALS geeft.

In hoofdstuk 3 hebben we onderzocht of er andere ziekten zijn die verband houden met ALS. Zo hebben we aangetoond dat patiënten met ALS vaker dan controlepersonen in de voorgeschiedenis een ernstig hoofdletsel hebben gehad, zoals een hersenschudding of schedelbasisfractuur (2.8% vs. 1.5%). Hoofdtrauma lijkt dus een mogelijke risicofactor voor ALS te zijn. Hierbij is het echter wel van groot belang om rekening te houden met de mogelijkheid van omgekeerde causaliteit, wat inhoudt dat het hoofdtrauma juist het gevolg zou kunnen van een beginnende maar nog onderkende ALS. Rekening houdend met deze eventuele omgekeerde causaliteit hebben we de analyse herhaald, waarbij hoofdtrauma opgelopen binnen vijf jaar voor de eerste symptomen van ALS geëxcludeerd werden. Ook dan blijkt er nog steeds een significant verband tussen hoofdtrauma en ALS, en blijft de hypothese, dat hoofdtrauma een risicofactor is voor ALS, staan. Bovendien wordt er in de literatuur steeds meer aandacht besteed aan de gevolgen van herhaald (licht tot matig) hoofdtrauma, zoals bij boksen, rugby, en ook bij voetbal gezien wordt. Er wordt beschreven dat deze sporters een verhoogd risico hebben om op de lange termijn geheugenstoornissen of andere centraal neurologische aandoeningen (zoals mogelijk ook ALS) te krijgen. Bij boksers met chronische traumatische encefalopathie worden tevens dezelfde pathologische afwijkingen in hersenen gevonden als bij ALS (TDP-43 inclusies), wat het verband tussen hoofdtrauma en ALS ondersteunt.

In eerdere onderzoeken is vaak gesuggereerd dat patiënten met ALS een gezonder cardiovasculair profiel hebben: zowel bij patiënten als bij hun familieleden komen cardiovasculaire ziekten minder vaak voor. Ook dit hebben wij in hoofdstuk 3 gestructureerd onderzocht. Het blijkt dat patiënten met ALS een lager cholesterol gehalte hebben in vergelijking met controlepersonen: ze rapporteren minder vaak hypercholesterolemie, 26% vs. 31%, en ze gebruiken beduidend minder statines, 12% vs. 21%. Dit sluit aan bij de bevinding dat patiënten gemiddeld een lagere body mass index (BMI) hebben, mediaan van 24 kg/m2 vs. 26 kg/m2. Een lager cholesterol gehalte en een lager BMI dragen bij aan een gezonder cardiovasculair risicoprofiel, maar dit lijkt op een


CHAPTER 10

152

meer indirecte relatie met ALS te berusten. We hebben daarnaast geen verband gevonden tussen ALS en auto-immuun aandoeningen, kanker of psychiatrische aandoeningen.

Voeding bevat allerlei bestanddelen die het risico op ziekten kunnen verhogen (zoals suikers en vetten) of juist kunnen beschermen tegen ziekten (zoals vitaminen). We hebben een grote studie uitgevoerd naar het verband tussen voedingspatronen en het risico op ALS (hoofdstuk 4). In samenwerking met Wageningen University is een gevalideerde voedsel-frequentie-vragenlijst afgenomen met meer dan 199 items. Uit de resultaten blijkt dat de totale hoeveelheid energie (in kilocalorieën, kCal) die patiënten tot zich nemen voordat zij ziek worden (2258 kCal), aanmerkelijk hoger is dan bij controlepersonen (2119 kCal). Daarnaast vinden we dat patiënten meer verzadigde vetzuren en cholesterol innemen dan controle personen.

Wanneer je bovenstaande in context tot eerder genoemde resultaten plaatst, lijkt er een verstoring in het evenwicht te zijn tussen input (verhoogde vet inname) en output (lager cholesterol gehalte en lagere BMI). Dit wijst mogelijk op een verstoord energie-metabolisme bij patiënten met ALS, waarbij ze meer calorieën verbranden in rust en bij beweging dan normaal, ook wel hypermetabolisme genoemd. Mogelijk wordt dit veroorzaakt door dysfunctie van de mitochondriën, de energiefabriekjes van het lichaam. De resultaten wijzen niet op een schadelijk effect van een hoogcalorisch dieet. Er is dus geen bezwaar tegen het toevoegen van extra calorieën aan de voeding als compensatiemechanisme voor het hypermetabolisme en om een daling van het lichaamsgewicht te beperken.

Nutriënten die een mogelijke beschermende werking zouden kunnen hebben, zijn onder andere glutamaat en antioxidanten, zoals vitamine C, vitamine E en lycopeen. We hebben echter geen verschil kunnen vinden in de inname van deze nutriënten tussen patiënten en controlepersonen. Wel hebben we in hoofdstuk 4 laten zien dat de inname van alcohol geassocieerd is met een lager risico op ALS. Eerdere studies suggereren een beschermend effect van specifiek rode wijn, welke antioxidanten bevat en mogelijk glutamaat geïnduceerde celdood kan tegen gaan. Wij hebben echter niet kunnen aantonen dat de relatie tussen alcohol en een verlaagd risico op ALS gedreven wordt door het drinken van rode wijn.

Deel II - OmgevingsfactorenIn het eerste hoofdstuk (hoofdstuk 5) van het deel over omgevingsfactoren, onderzoeken we beroepsmatige blootstelling aan potentieel schadelijke stoffen, zoals mineralen, dierlijke en plantaardige stoffen, pesticiden, gassen, metalen en oplosmiddelen. Hieruit blijkt dat hoge blootstelling aan diesel uitlaatgassen, zoals bij vrachtwagen- en buschauffeurs, militair personeel, mijnwerkers en spoorwegwerkers, een verhoogd risico op ALS geeft. Mogelijk dat inhalatie van de schadelijke uitlaatgassen zorgt voor een ontstekingsreactie van de zenuwcellen, dan wel voor oxidatieve stress, leidend tot ALS.


153

Nederlandse samenvatting

Behalve onderzoek naar blootstellingen vanuit het beroep, hebben we ook blootstelling aan gevaarlijke stoffen onderzocht vanuit de woonomgeving. Hiervoor hebben we historische adresgegevens verzameld van alle patiënten en controlepersonen via de gemeentelijke basisadministratie. Deze adresgegevens hebben we vervolgens in samenwerking met het Institute for Risk Assessment Sciences (IRAS) kunnen geocoderen en koppelen aan blootstellingskaarten van Nederland. Op deze manier hebben we in hoofdstuk 6 de relatie tussen blootstelling aan luchtvervuiling vanuit de woonomgeving en ALS op een gestructureerde manier kunnen onderzoeken. De gegevens over de luchtvervuiling concentraties zijn verzameld als onderdeel van een Europese studie, de European Study of Cohorts for Air Pollution Effects (ESCAPE). Uit ons Nederlandse onderzoek blijkt dat patiënten voorafgaand aan het ontstaan van de eerste symptomen gemiddeld een hogere blootstelling hadden aan fijnstof (specifiek fijnstof dat terug te vinden is in roetfilters; PM2.5absorbance) en stikstofoxiden (zoals NO2), in vergelijking met controlepersonen. Zelfs waarden van luchtvervuiling onder de Europese richtlijn vormen een risico voor ALS. Deze bevindingen onderbouwen tevens de relatie tussen beroepsmatige blootstelling aan diesel uitlaatgassen en een verhoogd risico op ALS zoals beschreven in hoofdstuk 5.

In hoofdstuk 7 hebben we de geografische gegevens gekoppeld aan een zeer gedetailleerde kaart van Nederland met daarop de precieze locaties van alle hoogspanningslijnen. De afstand tot een hoogspanningslijn staat gelijk aan de mate van blootstelling aan elektromagnetische straling. Voor elk woonadres van patiënten met ALS en controlepersonen hebben we met deze afstandsberekening de mate van blootstelling aan elektromagnetische straling vastgesteld. Uit de analyse blijkt dat er geen verschil is in afstand van de woning tot de hoogspanningslijnen tussen patiënten met ALS en controlepersonen. Blootstelling aan elektromagnetische straling vanuit de woonomgeving is dus geen grote risicofactor voor ALS.

Deel III - Genetische factorenIn het laatste deel van mijn proefschrift heb ik genetische risicofactoren voor ALS onderzocht. Hoofdstuk 8 beschrijft twee genen, hnRNPA1 en hnRNPA2B1, waarin DNA mutaties gevonden zijn in Amerikaanse families met multisysteem proteïnopathie (MSP). Dit is een zeldzame aandoening waarbij er degeneratie optreedt in verschillende orgaansystemen van het lichaam (hersenen, motorische zenuwen, spieren of botten) en er sprake is van karakteriserende TDP-43 pathologie. MSP is een overkoepelende naam voor symptomen die passen bij zowel inclusion body myositis (IBM: spierziekte), ziekte van Paget (botziekte), fronto-temporale dementie (FTD), als ook ALS. We hebben onderzocht of we in een Nederlandse populatie patiënten met enkel ALS (n=1219), FTD (n=142) of IBM (n=32) ook mutaties in deze genen konden vinden. Dit blijkt echter niet het geval, en ook vergelijkbare studies in Frankrijk en Italië rapporteren geen mutaties in deze genen. Hieruit hebben we geconcludeerd dat als mutaties in hnRNPA1 en hnRNPA2B1 al een risicofactor voor ALS vormen, dat deze dan in ieder geval zeldzaam zijn.


CHAPTER 10

154

In hoofdstuk 9 hebben we onderzocht of er vaker dan verwacht meerdere mutaties in verschillende bekende ALS genen in één persoon samen voorkomen. Uit eerder onderzoek is gebleken dat er bij patiënten met familiaire ALS vaker dan verwacht op basis van kans sprake is van meerdere mutaties. In de huidige studie hebben we dit ook onderzocht voor patiënten met sporadische ALS, waarbij de genetische predispositie minder duidelijk is. In deze studie hebben we in 4.1% van de patiënten met ALS en in 1.3% van de controlepersonen meerdere genetische mutaties gevonden. Dit is echter niet significant verschillend als je corrigeert voor de hogere toevalskans van ALS patiënten op één mutatie. Wel hebben we een significant verhoogde frequentie gevonden van patiënten die naast een C9orf72 repeat expansie nog een tweede variatie hebben in ATXN2, NIPA1 of SMN1. Specifiek de combinatie van C9orf72 en NIPA1 repeat expansies komen samen vaker voor dan je zou verwachten op basis van kans. Patiënten met een C9orf72 repeat expansie laten een grote diversiteit zien in klinische presentatie (zoals in leeftijd op het moment van de eerste symptomen, de lokalisatie van de eerste symptomen, of het al dan niet hebben van cognitieve klachten). Het is goed mogelijk dat een tweede genetische variatie (zoals een NIPA1 repeat expansie) bepaalt welk fenotype iemand ontwikkelt.

DiscussieIn hoofdstuk 10 vat ik de beschreven studies samen en plaats ik deze in een breder kader. Hierin bespreek ik onder andere dat ALS een complexe aandoening is waarbij een combinatie van leefstijlfactoren, omgevingsfactoren en genetische factoren een rol speelt in de pathogenese van ALS. Een recente Europese epidemiologische studie toont aan dat er een combinatie van zes verschillende stappen (factoren) nodig is om ALS te ontwikkelen. De complexiteit van de aandoening is ook terug te zien in de grote klinische diversiteit van ALS, zoals het verschil in leeftijd op het moment van de eerste symptomen (op elke volwassen leeftijd), de lokalisatie van de eerste symptomen (bijvoorbeeld zwakte van de armen of benen, spraak- of slikstoornissen), het al dan niet hebben van cognitieve klachten (in het kader van fronto-temporale dementie) en de ziekteduur. In plaats van het concept ALS als één ziekte zou je kunnen hypothetiseren dat deze diversiteit gezien moet worden in het kader van ALS als syndroom met verschillende subtypen. Verder onderzoek naar subtypen van ALS en combinaties van risicofactoren is nodig om meer inzicht te krijgen in deze dodelijke ziekte.


155

Dankwoord

DANKWOORD(Acknowledgements)

Zonder hoogte- en dieptepunten geen proefschrift. Gelukkig stond ik er tijdens mijn promotieonderzoek niet alleen voor. Ik wil hier graag een aantal mensen bedanken.

Allereerst alle ALS patiënten en familieleden: dank dat jullie allen jullie kostbare tijd hebben vrijgemaakt voor het onderzoek. Mijn waardering voor jullie inzet en medewerking in een voor jullie zo roerige tijd is groot. Natuurlijk wil ik daarbij ook alle controle personen bedanken die belangeloos aan het onderzoek naar ALS hebben meegewerkt.

Professor van den Berg, beste Leonard. Ik voel me nog steeds vereerd dat ik in jouw groep mijn promotieonderzoek heb mogen uitvoeren. Een groep die bloeit en groeit onder jouw duidelijke visie met als ultieme doel het ontwikkelen van nieuwe medicijnen voor ALS. Met jouw humor en enthousiasme weet jij mij en iedereen in de onderzoeksgroep te stimuleren tot onderzoek op wereldniveau.

Professor Veldink, beste Jan. Ik heb veel ontzag voor jouw diversiteit in kennis van zowel statistiek, epidemiologie als genetica. Jij kwam altijd met nieuwe ideeën en ingenieuze statistische oplossingen op de momenten dat ik het niet meer zag. En dat met een nooit aflatende en bewonderingswaardige gedrevenheid. Dank voor alle fijne samenwerking.

Dr. Vermeulen, beste Roel. Jij hebt met jouw kritische blik en frisse wind een waardevolle bijdrage geleverd om de studies naar omgevingsfactoren in relatie met ALS naar een hoger niveau te brengen. Jouw ervaring op dit gebied vanuit het IRAS en de samenwerking voor het koppelen van de data was onmisbaar voor de kwaliteit van het onderzoek.

Dr. van Es, beste Michael. Je weet mij (en vele anderen) altijd mee te nemen in jouw oneindig enthousiasme en ideeën over hoe we verder moeten komen in het ontrafelen van de genetische achtergrond van ALS en aanverwante neurologische aandoeningen. Een overredingskracht waar ik jaloers op ben.

Leden van de beoordelingscommissie, geachte prof. dr. F.L.J. Visseren, prof. dr. L.J. Kappelle, prof. dr. M.J.B. Taphoorn, prof. dr. ir. D.J.J. Heederik, prof. dr. P.I.W. de Bakker. Hartelijk dank dat ik u mijn proefschrift ter beoordeling mocht voorleggen.

Zonder de goede opzet van de PAN studie zou dit proefschrift en heel veel ander onderzoek in de ALS onderzoeksgroep niet mogelijk zijn geweest. Sonja, als eerste van de PAN onderzoekers, en daarna ook Mark, Perry en nu natuurlijk opgevolgd door Anne: bedankt voor alle inspanningen en zeer waardevolle brainstormmomenten.


CHAPTER 10

156

Gezien het succes van de PAN studie en de almaar groeiende dataset en onderzoeksprojecten is ook het ondersteunende PAN team in de tijd sterk uitgebreid. Petra en Hermieneke, toen ik begon met mijn promotieonderzoek waren jullie mijn steun en toeverlaat bij alle hobbels die we in de dataverzameling tegenkwamen, mijn dank daarvoor is groot. Maar ook de rest van het PAN team verdient het hier genoemd te worden. Met jullie hulp zal er zeker nog veel meer mooi onderzoek worden gerealiseerd.

Perry, Wouter en Kristel, dank voor jullie hulp en voornamelijk kennis ten aanzien van de genetica voor de laatste twee hoofdstukken van dit proefschrift. Altijd goed om te kunnen sparren. Peter, William, Raymond en Jelena, bedankt voor jullie ondersteuning bij het labwerk.

Alle andere labmatties: Oliver, Anna, Max, Ewout, Lotte, Renske, Frank, Henk-Jan, Gijs, Annelot, Marloes, Camiel, Dianne, Marc, Sandra. Ik heb veel geleerd van de soms verhitte (en veelal op luide toon gevoerde) discussies en natuurlijk ook veel lol gehad samen. Ik kan scooter inmiddels bijna waarderen, ik heb geen vliegangst meer en ik mis de koffietijd (en chocolade) op het lab nu al.

Nienke, zonder jou zou de MND poli lang niet zo goed lopen. Inmiddels regel jij de zaken daar niet meer alleen, maar nu samen met Kim. Keep up the good work, voor al die patiënten met ALS.

Annemarie, ook jij bent onmisbaar in alles wat je regelt en organiseert. Dank daarvoor.

Mijn paranimfen, Renée en Anne, ik ben zo blij dat jullie naast mij zullen staan op 16 juni. Renée, onze paden kruisen elkaar inmiddels zo vaak dat we ook echt niet meer om elkaar heen kunnen en gelukkig maar! Wat mij betreft hoeft dat nooit meer te veranderen. Anne, je bent begonnen als student bij mij en ik had me geen betere opvolging in het PAN onderzoek kunnen wensen. Als “PAN-dames” hebben we echt wel laten zien dat hard werken en lol hebben heel goed samen gaat.

Mijn collega’s van de afdeling neurologie van het MCH Westeinde. Ook al hebben jullie weinig direct met mijn promotieonderzoek te maken gehad, toch heb ik de laatste loodjes volbracht terwijl ik ben begonnen aan de opleiding tot neuroloog bij jullie. Ik voel me zeer op mijn plaats in de kliniek in Den Haag en ik vind het een voorrecht om deel van de groep uit te mogen maken.

Gerbrand, Renée, Joep, Bart en David, mede studenten vanaf jaar één. Niet iedereen zie ik nog even vaak, maar het voelt altijd goed om elkaar weer te zien en te spreken. Laten we dat zo af en toe vooral zo houden. Voor Gerbrand en Renée geldt het ‘af en toe’ gelukkig niet en ik hoop dat dat nog heel lang zo blijft en dat we nog eindeloos veel samen kunnen delen.


157

Dankwoord

Kirsten, Keetie, Rozemarijn, Marije, Nienke, Elleke, Lotte, Jenneke, Svetlana, Judith, Marjolijn, Renée, Perijne en alle anderen van het nog steeds groeiende en bloeiende medisch vrouwen dispuut Agnodice. Ik vind het heel bijzonder dat ik er vanaf het begin bij heb mogen zijn en dat we samen al zoveel lief en leed hebben kunnen delen tijdens de studie. De eerste huwelijken en baby (nog geen meervoud) hebben we inmiddels ook al gehad. Hopelijk volgen er nog vele mooie momenten.

Suzanne, Froukje (en alle andere teamgenootjes), we staan nu al weer heel wat jaren samen op het korfbalveld, al dan niet in hetzelfde team, maar altijd weten we elkaar weer te vinden. Zowel om het hoofd leeg te maken tijdens het sporten, danwel om achteraf onder een drankje alles te kunnen bespreken, en dat is echt heel fijn.

Fieke, dankzij onze moeders kennen we elkaar al sinds mijn geboorte. Inmiddels hebben wij al die jaren al een zeer waardevolle vriendschap. Ik ben heel blij dat we die ondanks onze veranderende levens nog steeds hebben.

De gehele schoonfamilie van Bergeijk-Lommerse, dank voor jullie betrokkenheid en interesse, iedere keer weer.

Pap, mam, Tessa en Daan, jullie onverwoestbare steun en geloof in mijn kunnen is mij zeer waardevol. Menige frustratie heb ik bij jullie kunnen uiten en weer loslaten. Een warmer thuis kan ik mij niet wensen. Gelukkig Daan, geef jij voor de afwisseling ook altijd de nodige nuchtere kritiek op het hele geneeskundige wereldje.

Tot slot, lieve Leendert, jij bent werkelijk mijn alles voor altijd.

Meinie.


CHAPTER 10

158

LIST OF PUBLICATIONS

In this thesis

Seelen M*, Huisman MHB*, de Jong SW, Dorresteijn KR, van Doormaal PT, van der Kooi AJ, de Visser M, Schelhaas HJ, van den Berg LH, Veldink JH. (2013) Lifetime time physical activity and the risk of amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry, Sep;84(9):976-81.

Seelen M, Visser AE, Overste DJ, Kim HJ, Palud A, Wong TH, van Swieten JC, Scheltens P, Voermans NC, Baas F, de Jong JM, van der Kooi AJ, de Visser M, Veldink JH, Taylor JP, van Es MA*, van den Berg LH*. (2014) No mutations in hnRNPA1 and hnRNPA2B1 in Dutch patients with amyotrophic lateral sclerosis, frontotemporal dementia and inclusion body myositis. Neurobiology of aging, Aug;35(8):1956.

Seelen M, Vermeulen RCH, van Dillen LS, van der Kooi AJ, Huss A, de Visser M, van den Berg LH*, Veldink JH*. (2014) Residential exposure to extremely low frequency electromagnetic fields and the risk of ALS. Neurology, Nov;83(19):1767-9.

Seelen M, van Doormaal PTC, Visser AE, Huisman MHB, Roozekrans MHJ, de Jong SW, van der Kooi AJ, de Visser M, Voermans NC, Veldink JH*, van den Berg LH*. (2014) Prior medical conditions and the risk of amyotrophic lateral sclerosis. J of Neurology, Oct;261(10):1949-56.

Other publications

Seelen M*, Cats EA*, Vlam L*, van Vught PW, van den Berg LH, van der Pol WL. (2011) Multifocal motor neuropathy is not associated with genetic variation in PTPN22, BANK1, Blk, FCGR2B, CD1A/E, and TAG-1 genes. J Pheriph Nerv Syst, Sep;16(3);179.

van Rheenen W, van Blitterswijk M, Huisman MH, Vlam L, van Doormaal PT, Seelen M, Medic J, Dooijes D, de Visser M, van der Kooi AJ, Raaphorst J, Schelhaas HJ, van der Pol WL, Veldink JH*, van den Berg LH*. (2012) Hexanucleotide repeat expansions in C9ORF72 in the spectrum of motor neuron diseases. Neurology, Aug;79(9):878 .

van Rheenen W, Diekstra FP, van Doormaal PT, Seelen M, Kenna K, McLaughlin R, Shatunov A, Czell D, van Es MA, van Vught PW, van Damme P, Smith BN, Waibel S, Schelhaas HJ, van der Kooi AJ, de Visser M, Weber M, Robberecht W, Hardiman O, Shaw PJ, Shaw CE, Morrison KE, Al-Chalabi A, Andersen PM, Ludolph AC, Veldink JH*, van den Berg LH*. (2013) H63D polymorphism in HFE is not associated with amyotrophic lateral sclerosis. Neurobiology of aging, May;34(5):1517.

Beeldman E, Jaeger B, Raaphorst J, Seelen M, Veldink J, van den Berg L, de Visser M, Schmand B. (2014) The verbal fluency index: normative data for cognitive testing in ALS. Amyotroph Lateral Scler Frontotemporal Degener, Sep;15(5-6):388-91.


159

Addendum

Dopper EGP, Seelen M, de Jong FJ, Veldink JH, van den Berg LH, van Swieten JC. (2013) Repeat-expansie in het C9orf72 gen: een link tussen frontotemporale dementie en amyotrofische laterale sclerose. Ned Tijdschr Geneeskd, 157:A6271.

Al-Chalabi A, Calvo A, Chio A, Colville S, Ellis CM, Hardiman O, Heverin M, Howard RS, Huisman MHB, Keren N, Leigh PN, Mazzini L, Mora G, Orrell RW, Rooney J, Scott KM, Scotton WJ, Seelen M, Shaw CE, Sidle KS, Swingler R, Tsuda M, Veldink JH, Visser AE, van den Berg LH, Pearce N. (2014) A Multistep model of amyotrophic lateral sclerosis. Lancet Neurology, Nov;13(11):1108-13.

Gallo V, Brayne C, Forsgren L, Barker RA, Petersson J, Hansson O, Lindqvist D, Ruffmann C, Ishihara L, Luben R, Arriola L, Bergareche A, Gavrila D, Erro ME, Vanacore N, Sacerdote C, Bueno-de-Mesquita HB, Vermeulen R, Seelen M, Sieri S, Masala G, Ramat S, Kyrozis A, Thricopolou A, Panico S, Mattiello A, Kaaks R, Teucher B, Katzke V, Kloss M, Curry L, Calboli F, Riboli E, Vineis P, Middleton L. (2014) Parkinson’s disease case ascertainment in the EPIC cohort: the NueroEPIC4PD study. Neurodegener Dis, accepted.

Submitted/in preparation

Huisman MHB, Seelen M, van Doormaal PTC , de Jong SW, de Vries JHM, van der Kooi AJ, de Visser M, Schelhaas HJ, van den Berg LH*, Veldink JH*. Presymptomatic BMI, and fat and alcohol consumption as independent risk factors for amyotrophic lateral sclerosis.

Seelen M*, Toro Campos RA*, Veldink JH, Visser AE, Hoek G, van der Kooi AJ, de Visser M, Raaphorst J, van den Berg LH#, Vermeulen RCH#. Long-term exposure to traffic related air pollution is associated with an increased risk of amyotrophic lateral sclerosis.

Seelen M, van Doormaal PTC, van Rheenen W, Bothof RJP, van Riessen T, Brands WJ, van der Kooi AJ, de Visser M, Voermans NC, Veldink JH, van den Berg LH#, van Es MA#. Large scale genetic screening in sporadic ALS identifies modifiers in C9orf72 repeat carriers.

Seelen, M*, Huisman MHB*, van Boxmeer L*, Visser AE, van Doormaal PTC, Raaphorst J, van der Kooi AJ, de Visser M, Vermeulen RCH, van den Berg LH, Veldink JH. Occupational exposure to diesel motor exhaust increases the risk of ALS.

Walhout R*, Schmidt R*, Westeneg HJ, Verstraete E, Seelen M, van Rheenen W, de Reus MA, Hendrikse J, Veldink JH, van den Heuvel MP#, van den Berg LH#. Effects of the C9orf72 repeat expansion: morphological changes in the brain of asymptomatic carriers and patients with amyotrophic lateral sclerosis.

*# These authors crontibuted equally to the manuscript


CHAPTER 10

160

ABOUT THE AUTHOR

Curriculum vitae

Meinie Seelen werd geboren in Arnhem op 12 januari 1986. Na het voltooien van het VWO aan het Arentheemcollege te Arnhem, studeerde zij geneeskunde aan de Universiteit Utrecht. Tijdens haar studie deed zij wetenschappelijk onderzoek in het Laboratorium voor Experimentele Neurologie in het UMC Utrecht naar multifocale motorische neuropathie. Hierna heeft zij gewerkt aan een extra-curriculair onderzoeksproject naar dopa-responsieve dystonie in het Skane University Hospital, in Lund (Zweden). Na haar artsexamen in 2010 heeft zij ervaring op gedaan als arts op de afdeling neurologie in de Tergooiziekenhuizen, te Blaricum. In 2011 begon zij aan haar promotieonderzoek onder leiding van prof. dr. L.H. van den Berg en prof. dr. J.H. Veldink, dat resulteerde in dit proefschrift. Per 2015 is zij begonnen aan de opleiding tot neuroloog in het MC Haaglanden.


Date post:	18-Jul-2020
Category:	Documents
Upload:	others
View:	4 times
Download:	0 times

Risk factors for Amyotrophic lateral sclerosis · Amyotrophic lateral sclerosis (ALS) is said to be...

Documents