Alzheimer's disease cerebrospinal fluid biomarkers in a cognitively healthy population and in patients with memory disorders : emphasis on co-morbidities and preanalytical factors

DISSERTATIONS | FANNI HAAPALINNA | ALZHEIMER’S DISEASE CEREBROSPINAL FLUID BIOMARKERS... | No 465

uef.fi

PUBLICATIONS OF

THE UNIVERSITY OF EASTERN FINLAND Dissertations in Health Sciences

ISBN 978-952-61-2799-6 ISSN 1798-5706

Dissertations in Health Sciences

PUBLICATIONS OF

THE UNIVERSITY OF EASTERN FINLAND

FANNI HAAPALINNA

ALZHEIMER’S DISEASE CEREBROSPINAL FLUID BIOMARKERS IN A COGNITIVELY HEALTHY POPULATION AND IN PATIENTS WITH MEMORY DISORDERS

Emphasis on co-morbidities and preanalytical factors Alzheimer’s disease (AD) cerebrospinal fluid

(CSF) biomarkers have been shown to reflect the neuropathological changes occurring in the brain already decades before the onset of the symptoms. However, there is a remarkable

overlap in the neuropathology between different neurodegenerative diseases, and both

co-morbidities and preanalytical factors may affect the results. This thesis sheds light on the accuracy of the CSF AD biomarkers in the

clinical diagnostics of memory diseases.

FANNI HAAPALINNA

UEF_Vaitoskirja_No_465_Fanni_Haapalinna_Terveystiede_kansi_18_05_23.indd 1 23.5.2018 12.45.31

Alzheimer’s disease cerebrospinal fluid biomarkers in a cognitively healthy population and in patients with memory

disorders

Emphasis on co-morbidities and preanalytical factors

FANNI HAAPALINNA

Alzheimer’s disease cerebrospinal fluid biomarkers in a cognitively healthy population and in patients with memory

disorders

Emphasis on co-morbidities and preanalytical factors

To be presented by permission of the Faculty of Health Sciences, University of Eastern Finland for public examination in the Auditorium Ca102, Canthia building in the University of Eastern Finland,

Kuopio, on Friday, September 7^th 2018, at 12 noon

Publications of the University of Eastern Finland Dissertations in Health Sciences

Number 465

Department of Neurology, Institute of Clinical Medicine, School of Medicine, Faculty of Health Sciences,

University of Eastern Finland Kuopio

2018

Grano Oy Jyväskylä, 2018

Series Editors:

Professor Tomi Laitinen, M.D., Ph.D.

Institute of Clinical Medicine, Clinical Physiology and Nuclear Medicine Faculty of Health Sciences

Associate Professor Tarja Kvist, Ph.D.

Department of Nursing Science Faculty of Health Sciences

Professor Kai Kaarniranta, M.D., Ph.D.

Institute of Clinical Medicine, Ophthalmology Faculty of Health Sciences

Associate Professor (Tenure Track) Tarja Malm, Ph.D.

A.I. Virtanen Institute for Molecular Sciences Faculty of Health Sciences

Lecturer Veli-Pekka Ranta, Ph.D. (pharmacy) School of Pharmacy

Faculty of Health Sciences

Distributor:

University of Eastern Finland Kuopio Campus Library

P.O.Box 1627 FI-70211 Kuopio, Finland http://www.uef.fi/kirjasto

ISBN (print): 978-952-61-2799-6 ISBN (pdf): 978-952-61-2800-9

ISSN (print): 1798-5706 ISSN (pdf): 1798-5714

ISSN-L: 1798-5706

III

Author’s address: Department of Neurology, Institute of Clinical Medicine, School of Medicine University of Eastern Finland

KUOPIO FINLAND

Supervisors: Sanna-Kaisa Herukka, M.Sc., M.D., Ph.D.

Department of Neurology, Institute of Clinical Medicine, School of Medicine University of Eastern Finland

KUOPIO FINLAND

Professor Anne Remes, M.D., Ph.D.

Department of Neurology, Research Unit of Clinical Neuroscience University of Oulu

OULU FINLAND

Reviewers: Adjunct Professor Susanna Melkas, M.D., Ph.D.

Clinical Neurosciences, Neurology

University of Helsinki and Helsinki University Hospital HELSINKI

FINLAND

Adjunct Professor Minna Raivio, M.D., Ph.D.

Speciality in Memory Diseases

Department of Geriatrics, School of Medicine University of Helsinki

HELSINKI FINLAND

Opponent: Adjunct Professor Mika H. Martikainen, M.D., Ph.D.

Faculty of Medicine, University of Turku

Division of Clinical Neurosciences, Turku University Hospital TURKU

FINLAND

IV

V

Haapalinna Fanni

Alzheimer’s disease cerebrospinal fluid biomarkers in a cognitively healthy population and in patients with memory disorders - Emphasis on co-morbidities and preanalytical factors

University of Eastern Finland, Faculty of Health Sciences

Publications of the University of Eastern Finland. Dissertations in Health Sciences Number 465. 2018. 65 p.

ISBN (print): 978-952-61-2799-6 ISBN (pdf): 978-952-61-2800-9 ISSN (print): 1798-5706 ISSN (pdf): 1798-5714 ISSN-L: 1798-5706

ABSTRACT

Neurodegenerative diseases are common, and represent a major public health challenge that is expected to become even greater in the future. The most common progressive neurodegenerative disease is Alzheimer’s disease (AD). Despite intense research, no cure has been found but with the medication currently available, it is possible to delay the progression of symptoms and support the functional ability of AD patients. The drug treatment has been shown to be most efficient during the earlier stages of the disease. Hence, early detection and differential diagnostics are crucial. Today, the diagnosis is based on a history of symptoms, clinical examination, neuropsychological evaluation, brain imaging and laboratory tests for differential diagnosis. The disease processes have been shown to begin decades before the first symptoms appear. Thus, multiple molecular biomarkers have been under development in order to aid in diagnostics. Currently, the only biomarkers in clinical use are cerebrospinal fluid (CSF) beta-amyloid 1-42 (Aβ1-42), tau and hyperphosphorylated tau (p-tau). These markers are included as supportive features in the criteria of AD. However, there is an overlap in both the clinical presentations and neuropathological changes between different neurodegenerative diseases. In addition, one out of every three cognitively healthy individuals has been shown to exhibit neuropathological changes typical for AD. These factors complicate making an accurate diagnosis. Furthermore, most previous research on CSF biomarkers has been conducted in academical centres that use highly standardized methods. Hence, it is possible that previous studies have overestimated the accuracy of the biomarkers in clinical practice.

The aim of this thesis was to investigate the associations between CSF AD biomarkers and cognition, and to evaluate the accuracy of the biomarkers in clinical diagnostics of memory diseases. In subjects with AD, poorer global cognitive performance was associated with higher levels of CSF tau and p-tau. On the contrary, in subjects with mild cognitive impairment (MCI), an association was found between low CSF Aβ1-42 and poorer performance in cognitive tests measuring Word List immediate and delayed recall. In addition, over half of the neurologically healthy subjects performed under the normal limit on one or more subtests of the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) battery. Twenty-eight percent of the healthy subjects also displayed AD-type pathological changes in their CSF, and these changes increased with age. The APOE-ε4 allele, which is recognized as the most significant genetic risk factor for sporadic AD, was associated with both decreased CSF Aβ1-42 and increased CSF tau concentrations. In this study, the sensitivities and specificities of the CSF AD biomarkers were lower than in most previous studies. However, sending the hospital CSF samples for analysis in sites other than the one in which the laboratory was located did not alter the accuracies of the markers.

Our results support previous findings that the best cognitive tests for detecting early AD pathological changes as well as low CSF Aβ1-42 concentration, are those measuring immediate and delayed recall. However, these cognitive changes as well as pathological changes in the CSF are common also in the neurologically healthy population, especially among the elderly. It is possible that these subjects are in the early stages of

VI

neurodegeneration representing preclinical AD. In everyday clinical diagnostics of memory diseases, the CSF AD biomarkers are not as accurate as has been previously claimed.

However, when combined, they may be used as a supportive tool, especially in unclear cases.

It is possible to reduce the effect of preanalytical factors on the accuracy by providing meticulous instructions on how best to collect and store the CSF samples.

National Library of Medicine Classification: WL 203, WL 358.5, WM 220, WT 155

Medical Subject Headings: Neurodegenerative Diseases; Dementia; Alzheimer’s Disease; ognitive Dysfunction;

Early Diagnosis; Aging; Cognition; Memory Disorders; Mental Status and Dementia Tests; Cerebrospinal Fluid;

Biomarkers; Apolipoprotein E4; Sensitivity and Specificity

VII

Haapalinna Fanni

Alzheimerin taudin aivo-selkäydinnestebiomarkkerit kognitiivisesti terveessä väestössä ja muistisairailla potilailla – Komorbiditeettien ja preanalyyttisten tekijöiden vaikutus

Itä-Suomen yliopisto, terveystieteiden tiedekunta

Publications of the University of Eastern Finland. Dissertations in Health Sciences Numero 465. 2018. 65 s.

ISBN (print): 978-952-61-2799-6 ISBN (pdf): 978-952-61-2800-9 ISSN (print): 1798-5706 ISSN (pdf): 1798-5714 ISSN-L: 1798-5706

TIIVISTELMÄ

Neurodegeneratiiviset sairaudet ovat merkittävä ja edelleen kasvava kansanterveydellinen haaste. Yleisin etenevä muistisairaus on Alzheimerin tauti (AT). Intensiivisestä tutkimuksesta huolimatta parantavaa hoitoa ei ole kyetty kehittämään. Nykyisillä lääkkeillä pystytään tukemaan AT-potilaan toimintakykyä ja hidastamaan oireiden etenemistä.

Hoidon on todettu olevan tehokkainta tautiprosessin varhaisessa vaiheessa. Tämän vuoksi etenevien muistisairauksien varhainen havaitseminen ja erotusdiagnostinen selvittely ovat olennaisessa asemassa. Nykyisin diagnoosi perustuu oirekuvaan, kliiniseen tutkimukseen ja neuropsykologiseen arvioon, aivokuvantamiseen sekä erotusdiagnostisiin laboratoriokokeisiin. Tautiprosessien on todettu alkavan vuosikymmeniä ennen ensimmäisten oireiden ilmenemistä. Diagnostiikan avuksi on pyritty kehittämään molekulaarisia biomarkkereita, joista tällä hetkellä kliinisessä käytössä ovat vain aivo- selkäydinnesteen beta-amyloidi 1-42 (Aβ1-42) sekä tau-proteiini ja hyperfosforyloitunut tau (p-tau). Edellä mainitut on hyväksytty tukeviksi piirteiksi AT:n diagnostisissa kriteereissä.

On kuitenkin todettu, että sekä neurodegeneratiivisten sairauksien kliinisissä kuvissa että patologisissa löydöksissä on päällekkäisyyttä. Lisäksi arviolta yhdellä kolmasosalla kognitiivisesti terveistä yksilöistä on todettavissa AT:lle tyypillisiä neuropatologisia muutoksia. Tämä vaikeuttaa sairauksien tarkan diagnoosin tekemistä. Lisäksi suurin osa aiemmista aivo-selkäydinnesteen biomarkkereiden tutkimuksista on tehty akateemisissa keskuksissa hyvin standardoiduilla menetelmillä, joten on mahdollista, että tutkimukset ovat siten yliarvioineet biomarkkeritestien tarkkuutta käytännön kliinisessä työssä.

Tämän väitöskirjatutkimuksen tavoitteena oli selvittää AT:n aivo-selkäydinnesteen biomarkkereiden yhteyttä kognition muutoksiin sekä niiden toimivuutta muistisairauksien käytännön kliinisessä diagnostiikassa. AT-potilailla todettiin yhteys yleisen kognition tason laskun ja selkäydinnesteen tau- ja p-tau-pitoisuuksien nousun välillä. Sen sijaan lievää kognitiivista heikentymää sairastavilla ja neurologisesti terveillä yksilöillä yhteys todettiin Aβ1-42:n pitoisuuden laskun ja sanalistan välitöntä ja viivästettyä palautusta mittaavien kognitiivisten testien tulosten huononemisen välillä. Lisäksi yli puolet neurologisesti terveistä yksilöistä suoriutuivat alle normaalin tuloksen rajan yhdessä tai useammassa Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) -tehtäväsarjan osa- alueessa. Terveistä yksilöistä 28%:lla oli todettavissa patologisia muutoksia selkäydinnesteen biomarkkereissa, ja muutokset lisääntyivät iän myötä. Sporadisen AT:n suurimmaksi geneettiseksi riskitekijäksi todettu APOE-ε4-alleeli oli yhteydessä sekä madaltuneeseen selkäydinnesteen Aβ1-42-pitoisuuteen että kohonneeseen tau-pitoisuuteen.

Retrospektiivisesti toteutetussa rekisteritutkimuksessa todettiin selkäydinnesteen AT- biomarkkereiden tarkkuuksien ja herkkyyksien olevan keskimäärin alhaisemmat kuin aiemmissa tutkimuksissa. Osuvuus ei kuitenkaan muuttunut, vaikka selkäydinnestenäyte oli lähetetty analysoitavaksi kaukaisemmasta sairaalasta analyysit suorittavan laboratorion yhteydessä olevan sairaalan sijaan.

Tutkimuksemme tukee aiempia löydöksiä, joissa kognitiivisista tutkimuksista parhaiten varhaisten AT-muutosten, alentuneen selkäydinnesteen Aβ1-42-pitoisuuden,

VIII

havaitsemisessa toimivat välitöntä ja viivästettyä muistiinpalautusta mittaavat testit. Edellä mainitut kognitiiviset muutokset sekä selkäydinnesteen biomarkkereiden patologiset pitoisuudet ovat kuitenkin yleisiä etenkin ikääntyneemmillä myös neurologisesti terveessä väestössä. On mahdollista, että nämä yksilöt sairastavat prekliinistä AT:a. Jokapäiväisessä muistisairauksien kliinisessä diagnostiikassa AT-biomarkkerit eivät ole yhtä tarkkoja kuin aiemmissa tutkimuksissa on todettu, mutta yhdessä ne toimivat diagnoosin tukena epäselvissä tapauksissa. Preanalyyttisten tekijöiden vaikutusta on mahdollista vähentää näytteenoton ja -säilytyksen tarkalla ohjeistuksella.

Luokitus: WL 203, WL 358.5, WM 220, WT 155

Yleinen suomalainen asiasanasto: neurodegeneratiiviset sairaudet; dementia; Alzheimerin tauti; ikääntyminen;

kognitio; diagnoosi; muistisairaudet; aivo-selkäydinneste; merkkiaineet; markkerit; apolipoproteiinit

IX

To my mother

X

XI

Acknowledgements

This study was carried out in University of Eastern Finland, Institute of Clinical Medicine – Neurology, and the Brain Research Unit of University of Eastern Finland during years 2013- 2018.

I wish to express my gratitude to my supervisors Sanna-Kaisa Herukka and Professor Anne Remes for the opportunity to participate in research in the Department of Neurology.

Thanks to Dr. Herukka M.Sc., M.D., Ph.D. for respecting my own views and for the trust she placed in me, as well as encouraging me to attend and present my results in international scientific conferences. Thanks to Professor Remes for her continuous support and sharing her knowledge. Thank you for the thought-provoking conversations which I believe have made me a better scientist. I admire your dedication. I also wish to express my gratitude to both of my supervisors for their understanding and support when I decided to go on an exchange for half a year, and become involved in somewhat time- consuming organizational activities.

I wish to thank the official reviewers of this thesis, Adjunct Professor Minna Raivio and Adjunct Professor Susanna Melkas from the University of Helsinki for constructive criticism and valuable comments that improved the final thesis.

I thank all the fellow co-authors and researchers with whom I had the opportunity to work. I am most grateful to my fellow doctoral student Olli Jääskeläinen from the University of Eastern Finland for his invaluable guidance with statistics, and not only for the interesting conversations but also for the too strong coffee we drank during the breaks. I want to express my gratitude to Teemu Paajanen Ph.D. from the Finnish Institute of Occupational Health especially for his valuable knowledge on the psychological aspect of neurodegenerative diseases. I also wish to thank Prof. Hannu Kokki, Docent Merja Kokki, Prof. Anne Koivisto, Prof. Hilkka Soininen, Docent Päivi Hartikainen, Merja Hallikainen M.D Ph.D., Docent Seppo Helisalmi, Eino Solje M.D.

Ph.D., Aku Kaipainen B.M., Eero Poukka B.M., Janne Penttinen M.D., Associate Prof.

Tuomo Hänninen, Maria Pikkarainen Ph.D., Virpi Moilanen M.D. Ph.D., Ilkka Tarvainen M.D., Pirjo Tuomainen M.D., Tero Tapiola M.D. Ph.D., Head Nurse Minna Rautiainen and senior laboratory technician Päivi Räsänen for their contribution.

I wish to express my gratitude to Kaisa Paldanius Ph.D., Professor Mikko Hiltunen and Associate Professor Annakaisa Haapasalo for introducing me to the world of science.

I thank Dr. Ewen MacDonald for proof-reading and correcting the language of the thesis.

I also owe a special thanks to Esa Koivisto for always helping me quickly with computer issues, even when I worked out of office. I want to thank Dr. Veli-Pekka Ranta for editing the thesis.

I want to express my warmest thanks to my dearest friends, most of whom I had the pleasure to get to know during the years spent working in the Finnish Medical Students’

International Committee. Niilo, Antti, Henri, Lari and Marianna, thank you for all the adventures and meetings. Laura, thank you for suggesting that I take a break and inviting me to go climbing or a dinner. Lassi and Joonas, thank you for all the support you have given me during these years. Simone and Laura, you made my exchange and proved that soul sisters exist. Maija, Inna, Roosa and Eskis, thank you for all the love and amazing moments together. Although we don’t see each other so often anymore, we keep in touch and I know you are always there for me. I look forward eagerly until we meet again. I also want to thank Hanna, Heidi and Anni as well as Milja and Tuomas. You were the best thing for me in Kuopio for the last two years, and I often miss you. Janne, thank you for the years we worked as a dream team, and all the experiences we have shared. Special thanks goes to Hanna and Janne with whom I’ve been able to share thoughts almost every day and who have without hesitation joined me in the craziest ideas, and Tuomas who has stayed by my side all the way from high school and with whom I can always laugh my head off.

XII

I owe my deepest gratitude to my mother Kirsi. Thank you for believing in me and for supporting me in my decisions even if you were afraid of what would happen. Thank you for always being there for me and offering your assistance, even with the figures of this thesis while not understanding their content. Thank you for everything. I want to thank my father Antti for all the support he has given me during these years. You have always been my role model as a researcher. I also wish to thank my dear sister Katri and her fiancé Niko for their support, especially in problems with my computer.

Finally, thank you Visa. You have believed in me more than I did myself, and lifted me up over and over again when I felt down and stressed. You were also the most valuable technical support for me during the writing process. Thank you for sharing quiet evenings, adventures, countless laughs and memorable moments with me, and reminding me of the most important things in life. I could not imagine sharing my life with anyone else. I love you more than words can ever tell.

I want to express my gratitude to the institutes and foundations that have been funding my research. These funding sources include Finnish Medical Society Duodecim and the Academy of Finland, Orion Research Foundation, Finnish Brain Foundation and Finnish Cultural Foundation’s North Savo Regional Fund as well as University of Eastern Finland.

Finally, I want to thank all the patients who participated in this study.

Rovaniemi, May 2018 Fanni Haapalinna

XIII

List of the original publications

This dissertation is based on the following original publications:

I Haapalinna F, Paajanen T, Penttinen J, Kokki H, Kokki M, Koivisto AM,

Hartikainen P, Solje E, Hänninen T, Remes AM, Herukka SK. Low Cerebrospinal Fluid Amyloid-Beta Concentration Is Associated with Poorer Delayed Memory Recall in Women. Dement Geriatr Cogn Dis Extra 19;6(2):303-12, 2016.

II Haapalinna F, Kokki M, Jääskeläinen O, Hallikainen M, Helisalmi S, Koivisto A, Paajanen T, Penttinen J, Pikkarainen M, Rautiainen M, Soininen H, Remes AM, Herukka S-K. Subtle cognitive impairment and Alzheimer’s disease-type pathological changes in cerebrospinal fluid are common among neurologically healthy subjects. Journal of Alzheimer’s disease 62(1): 165-174, 2018.

III Haapalinna F, Jääskeläinen O, Moilanen V, Tarvainen I, Tuomainen P, Kaipainen A, Poukka E, Nykänen N, Tapiola T, Remes AM, Herukka S-K. The Accuracy of Cerebrospinal Fluid Alzheimer’s Disease Biomarkers in Clinical Practice.

[Submitted]

The publications were adapted with the permission of the copyright owners.

XIV

XV

Contents

1 INTRODUCTION ... 1 2 REVIEW OF THE LITERATURE ... 2 2.1 Alzheimer’s disease ... 2 2.1.1 Epidemiology ... 2 2.1.2 Clinical features ... 2 2.1.3 Diagnostics... 3 2.1.4. Pathophysiology ... 6 2.1.5 Genetic background ... 8 2.1.6 Other risk factors ... 9 2.2 Other dementing disorders ... 9 2.2.2 Frontotemporal Lobar Degeneration ... 9 2.2.2 Vascular dementia ... 10 2.2.3 Dementia with Lewy bodies ... 10 2.2.4 Creutzfeldt-Jakob disease ... 11 2.2.5 Normal pressure hydrocephalus ... 11 2.3 Mild cognitive impairment ... 11 2.4 Apolipoprotein E E4 ... 13 2.5 Consortium to Establish a Registry for Alzheimer’s Disease Neuropsychological Battery ... 13 2.6 Molecular biomarkers for neurodegenerative diseases ... 15 2.6.1 CSF Aβ1-42, tau and p-tau ... 16 2.6.2 Other biomarkers ... 18 2.6.3 Challenges ... 20 3 AIMS OF THE STUDY ... 22 4 MATERIALS AND METHODS ... 23 4.1 Study populations ... 23 4.1.1 Study I ... 23 4.1.2 Study II ... 23 4.1.2 Study III ... 23 4.2 Assessment of cognition ... 24 4.2.1 Consortium To Establish a Registry for Alzheimer’s Disease Neuropsychological Battery ... 24 4.2.2 TELE phone call cognitive test (study II) ... 24 4.3 CSF samples and biomarker analyses ... 25 4.4 APOE analyses (Study II) ... 26 4.5 Statistical analyses ... 26 4.6 Ethical aspects ... 26 5 RESULTS ... 27 5.1 CSF AD biomarkers and cognition in subjects with AD, MCI and no subjective cognitive symptoms (Studies I and II) ... 27 5.2 AD-type CSF pathology and cognitive impairment among neurologically healthy subjects (Study II) ... 29

XVI

5.3 The effect of APOE-ε4 allele on CSF biomarkers and cognition (Study II) ... 30 5.4 The effects of hypercholesterolemia, cardiovascular diseases, hypertension and diabetes on CSF biomarkers and cognition (study II) ... 31 5.5 There are no significant differences in the accuracy of the CSF AD biomarker measurements between centers in everyday clinical diagnostics (Study III) ... 31

5.5.1 Intercenter variability ... 31 5.5.2 Differences between diagnostic groups ... 33 6 DISCUSSION ... 35 6.1 Associations between the CSF AD biomarkers and cognition (studies I and II) ... 35 6.2 AD-type CSF pathology and cognitive impairment among neurologically healthy subjects (Study II) ... 36

6.2.1 CSF AD biomarkers ... 36 6.2.2 Performance on cognitive tests ... 36 6.2.3 Associations between CSF AD biomarkers and cognition ... 37 6.3 APOE-ε4 allele and CSF biomarkers (Study II) ... 37 6.4 Hypercholesterolemia, cardiovascular diseases and cognition (Study II) ... 38 6.5 Clinical variability of CSF AD biomarker analyses (Study III) ... 38 6.6 Strengths and limitations of the study ... 40 7 CONCLUSIONS ... 42 8 FUTURE PERSPECTIVES ... 43 REFERENCES ... 44

XVII

Abbreviations

AD Alzheimer’s disease

ABCA7 ATP-binding cassette, subfamily A, member 7 AIC Amyloid precursor protein intracellular domain

ALS Amyotrophic lateral sclerosis

ANCOVA Analysis of covariance

ANOVA Analysis of variance

APOE-ε4 Apolipoprotein E ε4 allele

APP Amyloid precursor protein

Aβ Beta-amyloid

BBB Blood-brain-barrier

BIN1 Bridging integrator 1

BIOMARKAPD Biomarkers for Alzheimer’s and Parkinson’s disease

BNT-15 15-item Boston Naming Test

bvFTD Behavioral variant frontotemporal degeneration

CAA Cerebral amyloid angiopathy

CADASIL Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy

CERAD-NB Consortium to Establish a Registry for Alzheimer’s Disease- Neuropsychological Battery

CJD Creutzfeld-Jacob’s disease

CLU Clustering/Apolipoprotein J

CR1 Complement receptor 1

CSF Cerebrospinal fluid

CT Computed tomography

C9ORF72 Chromosome 9 open reading frame 72

DLB Dementia with Lewy bodies

DSM-V Diagnostic and Statistical Manual of Mental Disorders fifth edition FDG-PET Fluorodeoxyglucose positron emission tomography

FTLD Frontotemporal lobar degeneration

GRN Progranulin

IWG International Working Group

KUH Kuopio University Hospital

LPA Logopenic progressive aphasia

l-vPPA Logopenic-variant primary progressive aphasia MAPT Microtubule associated protein tau

MCI Mild cognitive impairment

XVIII

miRNA Micro ribonucleic acid

MMSE Mini mental state examination

MRI Magnetic resonance imaging

MTA Medial temporal lobe atrophy

MTS Memory Total Score

NINCDS-ADRDA National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association

NFH Neurofilament heavy protein

NFL Neurofilament light protein

NFT Neurofibrillary tangle

nfvPPA Nonfluent variant primary progressive aphasia NOTCH3 A gene coding notch homolog 3 protein

NPH Normal pressure hydrocephalus

NSE Neuron specific enolase

OUH Oulu University Hospital

PCA Posterior cortical atrophy

PD Parkinson’s disease

PiB Pittsburgh compound B

PPA Primary progressive aphasia

PrP^C Cellular prion protein

PSEN1+2 Presenilin 1 and 2

p-tau Hyperphosphorylated tau

REM Rapid eye movement

RNA Ribonucleic acid

ROC Receiver Operating Characteristic

SMC Subjective memory complaint

svPPA Semantic variant primary progressive aphasia

TDP-43 TAR DNA-binding protein 43

TDM Transmembrane domain

TLR Toll-like receptor

VaD Vascular dementia

VCI Vascular cognitive impairment

VLP-1 Visinin-like protein 1

1 Introduction

Cognitive impairment and dementing diseases are common, and due to the increase in life expectancy, their prevalence is expected to increase significantly in the future. These illnesses represent a significant social problem from the individual patient and his/hers loved ones as well as a financial burden on the economy (Wimo, Jonsson et al. 2013).

Dementia is a syndrome in which a group of symptoms affect multiple cognitive domains so severely to impair activities of daily functioning (McKeith, Cummings 2005). In 2010, there were 36.6 million people suffering from dementia all around the world, most of them in the developing countries. It has been estimated that the number of sufferers will double every 20 years and accordingly by 2050, as many as 115 million people will be diagnosed with dementia (Prince, Bryce et al. 2013).

The most common cause of dementia is Alzheimer’s disease (AD), accounting for approximately 70% of all cases (Reitz, Mayeux 2014, Mayeux, Stern 2012a). Other common causes include vascular dementia (VaD), dementia with Lewy bodies (DLB) and Frontotemporal lobar degeneration (FTLD). Furthermore, other etiologies such as excessive alcohol consumption and head trauma may lead to cognitive impairment and eventually dementia (Mayeux, Stern 2012b).

There is an overlap between different neurodegenerative diseases in terms of both their genetic background and pathological processes as well as their clinical picture. One common feature is the misfolding and accummulation of fibrillary protein aggregates (Lausted, Lee et al. 2014), such as beta-amyloid (Aβ) in AD (Reitz, Mayeux 2014). It has been proven that the pathological processes in the brain begin already decades prior to the appearence of any clinical symptoms (Price, Morris 1999a, Blennow, Dubois et al. 2015). However, the same kind of neuropathology can be observed also in some individuals who remain cognitively intact (De Meyer, Fred Shapiro et al. 2010, Knopman, Parisi et al. 2003a).

The diagnosis of dementing diseases is typically based on the clinical course of the symptoms, a clinical examination, neuropsychological testing, brain imaging and differential laboratory tests. However, at the present time, a definite diagnosis can be achieved only in a post-mortem neuropathological examination or after a brain biopsy. A number of biomakers have been investigated in order to aid in the diagnosis and differential diagnosis of neurodegenerative diseases, as well as to follow the progression of the diseases and the effects of therapeutic interventions. To date, there are validated disease-specific molecular biomarkers only for AD (Lausted, Lee et al. 2014).

At the present time, there is no curative treatment for neurodegenerative diseases, but it is possible to treat the symptoms of the disease (Carter, Simms et al. 2010). This is most efficiently done in the earlier stages of the pathology. In the future, if one could identify reliable neuropathology-specific cerebrospinal fluid (CSF) biomarkers, then the diagnosis could be achieved already in the disease’s prodromal phase.

2

2 Review of the Literature

2.1 ALZHEIMER’S DISEASE

2.1.1 Epidemiology

Alzheimer’s disease is the most common cause of dementia accounting for approximately 70% of all cases (Reitz, Mayeux 2014, Querfurth, LaFerla 2010). In 2006, it was estimated that there were around 26.6 million patients with AD (Brookmeyer, Johnson et al. 2007), with the actual numbers being highest in countries with low and middle incomes. In relation to the size of the population, the prevalence is, however, higher in Northern America and Western Europe. As with dementia in general, the prevalence increases almost exponentially with age and is expected to double every 20 years (Reitz, Mayeux 2014, Mayeux, Stern 2012a, Ballard, Gauthier et al. 2011). It has been estimated that in 2050, there will be more than 100 million people affected by AD (Brookmeyer, Johnson et al. 2007).

Women are predominantly affected by AD; it has been claimed that women account for approximately two thirds of all AD cases (Launer, Andersen et al. 1999, Andersen, Launer et al. 1999).

2.1.2 Clinical features

The prodromal phase of the disease precedes the AD dementia syndrome; at this time, a certain degree of cognitive impairment can be detected together with AD-type pathological changes in either imaging or CSF (Pena-Casanova, Sanchez-Benavides et al. 2012). The first symptom is typically an impairment of episodic memory that is also the most characteristic feature of AD (Dubois, Feldman et al. 2014, Salmon, Bondi 2009) occurring in as many as 94%

of all patients (Dubois, Feldman et al. 2014). The actual AD typically progresses in three stages: from mild to moderate and eventually to severe dementia. In mild AD, in conjuction with memory disturbances also other areas of cognition, especially executive functions and language skills, become affected. In addition, agnosia and apraxia are common. The patients start to experience problems in more complex activities of daily living, such as financial tasks.

In the moderate stage of the disease, the impairment of cognition becomes more severe and challenges in daily living are more apparent. In addition to losing older memories, orientation with respect to time and space may be impaired, and there is an increased risk for wandering and becoming lost. The patient starts to require a greater level of care. In the final, severe, stage of the disease, cognition continues to worsen. The patients struggle with communication, and eventually lose their ability to respond to the environment and control movement, even swallowing (McKhann, Knopman et al. 2011, Dubois, Feldman et al. 2007, Weintraub, Wicklund et al. 2012, Bondi, Jak et al. 2008, Blennow, de Leon et al. 2006). In addition to cognitive impairments, neuropsychiatric symptoms are common in AD, affecting up to 90% of the patients at some point in their disease course (Robert, Verhey et al. 2005, Lyketsos, Carrillo et al. 2011). Some of these symptoms may be present already in the prodromal phase of the disease, and they have also been proven to increase the risk of developing AD as well as accelerating the progression from mild cognitive impairment (MCI) into dementia (Geda, Roberts et al. 2008). Apathy is the most common of the symptoms, and it typically persists throughout the disease course (Lyketsos, Lopez et al.

2002, Robert, Verhey et al. 2005, Devanand, Jacobs et al. 1997). Depression and alterations in sleep-wake rhythm are also frequently seen, and are present already in early stages of the disease. As the disease progresses, more severe neuropsychiatric symptoms appear. The patients may start to exhibit aggression as well as suffering delucions or hallucinations, even

3

outright psychosis is possible (Lyketsos, Carrillo et al. 2011, Robert, Verhey et al. 2005).

However, it should be borne in mind that the disease progression regarding both symptoms and survival varies between individuals (Brookmeyer, Johnson et al. 2007, Frisoni, Rozzini et al. 1999, Lopez, Becker et al. 2003). The mean survival after the diagnosis has been estimated to be approximately six years, but it is dependent on the time at which the diagnosis is achieved, the subtype of the disease, and on the presence of possible comorbidities. If diagnosed early, the survival of the patient may be as long as 30 years.

(Brookmeyer, Johnson et al. 2007, Armstrong 2014, van de Vorst, Vaartjes et al. 2015).

Aspiration pneumonia is the most common cause of death in AD patients (van der Steen, Ooms et al. 2002).

In addition to typical amnestic AD, atypical variants have been observed. They are usually divided into posterior cortical atrophy (PCA), logopenic progressive aphasia (LPA) and a frontal variant of AD (Dubois, Feldman et al. 2014, Galton, Patterson et al. 2000, Wolk 2013).

These atypical presentations have been shown to be rather common, especially among younger AD subjects (Smits, Pijnenburg et al. 2012).

2.1.3 Diagnostics

Today, the certain diagnosis of AD can only be achieved by brain biopsy. Clinical diagnostics is based on clinical and neuropsychological evaluation, brain imaging, and laboratory findings typical of the disease or findings excluding other reasons behind the symptoms. The AD diagnostic criteria according to Dubois et al. (Dubois, Feldman et al. 2007) are depicted in Table 1.

A typical finding in neuropsychological testing is poor performance in tests measuring delayed recall. Furthermore, the benefit of cueing is reduced in AD patients (Dubois, Feldman et al. 2007). In order to be diagnostical for AD, the impairment of episodic memory has had to be seen early in the course of the symptoms and to have progressed for at least six months (Dubois, Feldman et al. 2007).

In structural neuroimaging (magnetic resonance imaging (MRI) and computed tomography (CT)), the most pronounced brain atrophy is located in the medial temporal structures (entorhinal cortex, hippocampus, parahippocampal gyrus) already during the early stages of the disease but global brain atrophy may also be seen (Dubois, Feldman et al.

2007). In addition to structural brain imaging, functional imaging such as fluorodeoxyglucose positron emission computed tomography (FDG-PET) may be used. In those procedures, bilateral hypoperfusion and hypometabolism are detected in the temporal and parietal cortices (Hoffman, Welsh-Bohmer et al. 2000, Dubois, Feldman et al. 2007). In addition, amyloid-PET may be useful. For example, it has been proven that the radiotracer Pittsburgh compound B (PiB) binds highly selectively to Aβ deposits in the brain (Ikonomovic, Klunk et al. 2008), which enables the evaluation of the amyloid burden with PET imaging. The meta-analysis conducted by Ossenkoppele et al. (Ossenkoppele, Jansen et al. 2015) found that the prevalence of amyloid in a PET scan decreased with age in AD subjects, whereas the prevalence increased with age in most non-AD dementia subjects.

Hence, amyloid PET has been postulated as being helpful in the differential diagnostics of dementias, especially in the early stages of the diseases (Ossenkoppele, Jansen et al. 2015).

Certain CSF biomarkers specific for AD have also been identified. They reflect the pathological changes in the brain in an opposite manner, leading to reduced levels of Aβ1- 42 and, on the contrary, increased levels of tau and p-tau in the CSF (Mattsson, Blennow et al. 2009, Tapiola, Alafuzoff et al. 2009a). The pathological changes in the brain begin already 20-30 years prior to the appearance clinical symptoms (Blennow, Dubois et al. 2014, Price, Morris 1999b), as do the changes in the levels of these biomarkers in the CSF (Figure 1) (Jack, Knopman et al. 2013). Hence, the CSF biomarkers make possible a very early diagnosis of AD. The biomarkers will be discussed further in chapter 2.7.

4

Figure 1. Model of dynamic biomarkers of the pathological cascade of Alzheimer’s disease pathology. Modified from Jack et al. 2013.

In addition to fulfilling the core diagnostic criteria and supportive features, other possible reasons, such as depression, which could account for the memory deficit, must be excluded (Dubois, Feldman et al. 2007). Despite continuous revision, the clinical criteria for diagnosing AD are still considered as imperfect (Beach, Monsell et al. 2012).

Table 1. The National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease (NINCDS-ADRDA) diagnostic criteria of AD according to Dubois et al. 2007.

Probable AD: A plus one or more supportive features B, C, D or E Core diagnostic criteria

A. Presence of an early and significant episodic memory impairment that includes the following features:

1. Gradual and progressive change in memory function reported by patients or informants over more than 6 months

2. Objective evidence of significantly impaired episodic memory on testing: this generally consists of recall deficit that does not improve significantly or does not normalize with cueing or recognition testing and after effective encoding of information has been previously controlled

3. The episodic memory impairment can be isolated or associated with other cognitive changes at the onset of AD or as AD advances

5

Table 1. Continues.

Supportive features

B. Presence of medial temporal lobe atrophy

 Volume loss of hippocampi, entorhinal cortex, amygdala evidenced on MRI with qualitative ratings using visual scoring (referred to well characterized population with age norms) or quantitative volumetry of regions of interest (referenced to well characterized population with age norms)

C. Abnormal cerebrospinal fluid biomarker

 Low amyloid β1-42 concentration, increased total tau concentration, or increased phospho- tau concentration, or combinations of the three

 Other well validated markers to be discovered in the future D. Specific pattern on functional neuroimaging with PET

 Reduced glucose metabolism in bilateral temporal parietal regions

 Other well validated ligands, including those that foreseeably will emerge such as Pittsburgh compound B or FDDNP

E. Proven AD autosomal dominant mutation within the immediate family Exclusion criteria

History

 Sudden onset

 Early occurrence of the following symptoms: gait disturbances, seizures, behavioural changes Clinical features

 Focal neurological features including hemiparesis, sensory loss, visual field deficits

 Early extrapyramidal signs

Other medical disorders severe enough to account for memory and related symptoms

 Non-AD dementia

 Major depression

 Cerebrovascular disease

 Toxic and metabolic abnormalities, all of which may require specific investigations

 MRI FLAIR or T2 signal abnormalities in the medial temporal lobe that are consistent with infectious vascular insults

Criteria for definite AD

AD is considered definite if the following are present:

 Both clinical and histopathological (brain biopsy or autopsy) evidence of the disease, as required by the NIA-Reagan criteria for the post-mortem diagnosis of AD; criteria must both be present

 Both clinical and genetic evidence (mutation on chromosome 1, 14 or 21) of AD; criteria must both be present

There is extensive mixed neuropathology between AD and other dementing diseases (Ewers, Mattsson et al. 2015) and, furthermore, it has been shown that as many as every third of cognitively healthy elderly person displays an AD-type pathology in their brain (Hulette, Welsh-Bohmer et al. 1998, Knopman, Parisi et al. 2003a). In addition, AD-type pathological changes can be found in the brain of most subjects with DLB (Schneider, Arvanitakis et al.

2007). Similarly, up to 70% of AD cases have been reported to display Lewy body pathology in their brain (Walker, McAleese et al. 2015).

6

2.1.4. Pathophysiology Amyloid pathology

The main feature of AD pathology is the formation and accumulation of amyloid plaques in the brain (Braak, Braak 1991). According to the amyloid cascade hypothesis Aβ-plaques begin to form early in the disease process, preceeding tau pathology (Selkoe, Hardy 2016).

The amyloid plaques consist mostly of Aβ-peptides formed during the sequential cleavages of amyloid precursor protein (APP). In the non-amyloidogenic pathway, amyloid precursor protein (APP) is first cleaved by α- and then by γ-secretase (Figure 2). In the amyloidogenic pathway, APP is targetted by β-secretase, after which a cleavage by γ- secretase leads to the formation of Aβ-peptides. There are different species of Aβ but the most common ones contain 40 (Aβ40) and 42 (Aβ42) amino acids. Most of the diffuse amyloid accumulating in the brain consists of Aβ40, whereas Aβ42 is more prone to form insoluble oligomers. Monomeric Aβ peptides have not been proved to be toxic towards neurons, but after forming oligomers, the Aβ evokes oxidative stress and impairs the function of the neurons and synapses, eventually leading to the death of affected neuronal cells (Ballard, Gauthier et al. 2011). However, it is noteworthy that diffuse plaques are also encountered in normal aging (Delacourte, David et al. 1999, Hulette, Welsh-Bohmer et al. 1998, Knopman, Parisi et al. 2003b).

Figure 2. Amyloidogenic processing of the amyloid precursor protein (APP) and generation of the Aβ peptides. Aβ: amyloid-β peptide; AICD: amyloid precursor protein intracellular domain; sAPPβ:

soluble fragment amyloid-β peptide; TMD: transmembrane domain. The large arrow represents the accumulation of amyloid plaques. Modified from Bekris er al (Bekris, Yu et al. 2010).

In AD, either the production of Aβ is increased or its clearance from the brain is impaired, since both processes can lead to increased accumulation of amyloid plaques (Ballard, Gauthier et al. 2011). This abnormal Aβ turnover reduces the level of of Aβ in the CSF. The glymphatic system of the brain has been suggested to be involved in the circulation of Aβ (Iliff, Wang et al. 2012, Louveau, Da Mesquita et al. 2016). In particular, the choroid plexus has been proposed to play a crucial role (Alvira-Botero, Carro 2010); furthermore the clearance of Aβ through this pathway appears to be more efficient during sleep (Xie, Kang et

7

al. 2013). However, neither the role nor the mechanism of how the glymphatic system affects the circulation of Aβ in AD are fully understood.

In typical AD, the accumulation sites of Aβ plaques can be divided into five phases (Figure 3). In the first phase, the frontal, parietal, temporal and occipital lobes are involved. In the second phase, the deposits spread into the entorhinal and insular cortices. In phase three, the plaques can also be detected in the striatum and diencephalic nuclei. The fourth phase is considered to be achieved when amyloidosis reaches the nuclei in the brainstem (substantia nigra, red nucleus, central gray, superior and inferior follicle, inferior olivary nucleus, and intermediate reticular zone). Finally, in the fifth phase the Aβ deposits have spread into cerebellum and additional brainstem nuclei (locus coeruleus, pontine nuclei, parabrachial nuclei, reticulo-tegmental nucleus, dorsal tegmental nucleus, and oral and central raphe nuclei)(Thal, Rub et al. 2002, Alafuzoff, Thal et al. 2009).

Figure 3. Phases of β-amyloid accumulation. Neocortex: black. The arrows in each picture depict the brain areas that become affected in that phase. Modified from Thal, Rub et al. 2002.

Tau pathology

In its physiological forms and concentration, tau is a protein that supports the microtubules present in the axons of neurons as well as possesses a role in axonal transport. The pathological form of tau is produced by hyperphosphorylation, leading to its aggregation into neurofibrillary tangles (NFTs) (Lee, Goedert et al. 2001, Ballatore, Lee et al. 2007). This hyperphosphorylation of tau impairs axonal function and makes the neurons more vulnerable to oxidative stress (Ballatore, Lee et al. 2007). With neuronal death and axonal damage, tau and p-tau are released into the CSF, hence leading to an increase in their concentration in the fluid (Tapiola, Alafuzoff et al. 2009b). This is, however, not specific for AD but also occurs in other events causing neuronal damage (Lee, Goedert et al. 2001).

The accumulation of NFTs is divided into six stages (Figure 4). In stages I and II, NFTs are seen in the transentorhinal region, whereas in stages III and IV, also entorhinal and later hippocampal areas become affected while the involvement of the transentorhinal region becomes considerable. As the disease progresses, NFTs start to involve the neocortical areas of the brain. This is considered as stages V and VI. The clinical symptoms of AD begin to appear in stages III and IV (Braak, Braak 1991).

8

Figure 4. Distribution of neurofibrillary (NF) changes in Alzheimer’s disease. The increasing density of shading indicates the increasing severity of pathological NF changes. Modified from Braak & Braak 1991.

2.1.5 Genetic background

Mutations in genes coding for presenilin 1 (PSEN1) and 2 (PSEN2) or APP are associated with young onset autosomal dominant AD. These mutations are very rare, and account for only 1% of AD cases, with mutations in PSEN1 being most common (20-70%). Mutations in APP account for 10-15% of the total number of mutations, whereas those in PSEN2 are rare (Bekris, Yu et al. 2010). All these mutations cause an increased processing of APP by the pathway producing more Aβ1-42 than soluble Aβ1-40. Hence, the disease process begins earlier, the progression is rapid, and the symptomatic phase is reached at an early age before 65 years, even before an individual’s 30^th birthday (Bekris, Yu et al. 2010).

Late onset AD manifesting after 65 years of age, on the contrary, is responsible for over 90% of all AD cases (Bekris, Yu et al. 2010). It has been shown to have an inheritability of 59- 78% (Gatz, Reynolds et al. 2006). The most common genetic risk factor is APOE-ε4 allele (Bekris, Yu et al. 2010). One copy of the allele increases the risk of AD by threefold, and two copies by up to 12-fold as compared to ε4 negative subjects (Holtzman, Herz et al. 2012).

APOE-ε4 will be described in more detail in chapter 2.5. Many other risk gene candidates have also been identified. The most widely studied mutations after APOE according to AlzGene database (www.alzgene.org) include Bridging integrator 1 (BIN1), Clusterin/Apolipoprotein J (CLU), ATP-binding casette subfamily A member 7 (ABCA7) and Complement receptor 1 (CR1). The majority of these other risk genes pose a minor risk, but when present together, they may increase the risk considerably (Ballard, Gauthier et al. 2011).

However, the genetics is complex, and despite an increased risk, the carriers of these mutations may not necessarily develop AD. (Bekris, Yu et al. 2010, Gatz, Reynolds et al. 2006, Mez, Marden et al. 2017).

9

2.1.6 Other risk factors

The most important risk factor for AD is age. The prevalence of the disease has been shown to increase exponentially after the age of 65 years, and after 85 years of age, up to one third of the population can be diagnosed with AD (Querfurth, LaFerla 2010, Reitz, Mayeux 2014).

In addition, it has been proven that various lifestyle, socioeconomical and environmental factors increase the risk for dementia. These factors include, e.g., low educational level and excessive alcohol consumption. In addition, factors associated with vascular diseases, such as hypertension, obesity and smoking have been shown to increase the risk (Norton, Matthews et al. 2014, Barnes, Yaffe 2011, Reitz, Mayeux 2014).

On the other hand, cognitive reserve comprising of many factors such as high educational level and mental activity, as well as a healthy lifestyle including consuming a Mediterranean diet and participating in physical activity have been shown to be beneficial. In addition, social networks have been postulated to protect from AD (Patterson, Feightner et al. 2007, Reitz, Mayeux 2014, Blennow, de Leon et al. 2006, Ballard, Gauthier et al. 2011, Wilson, Scherr et al.

2007). However, some studies have found no association between later life lifestyle factors and conversion to AD (Reijs, Vos et al. 2017).

2.2 OTHER DEMENTING DISORDERS

2.2.2 Frontotemporal Lobar Degeneration

Frontotemporal Lobar Degeneration is a heterogenous group of neurodegenerative syndromes manifesting in progressive behavioral symptoms, problems with language or motor dysfunction (Neary, Snowden et al. 1998). It is the cause of dementia in approximately 5-10 % of cases. FTLD usually begins before the age of 65 years, and in under 65-year-olds, it accounts for 10-20 % of all dementia cases, making it the second common cause of early onset dementia (Ratnavalli, Brayne et al. 2002, Neary, Snowden et al. 2005).

FTLD typically affects both the frontal and temporal lobes of the brain (Cairns, Bigio et al.

2007). Based on the dominating symptoms at the beginning of the disease, FTLD is divided into two major clinical syndromes: behavioral variant frontotemporal dementia (bvFTD) and primary progressive aphasia (PPA) (Rascovsky, Hodges et al. 2011). BvFTD is the most common form of FTLD accounting for two out of every three cases (Johnson, Diehl et al.

2005). It is characterized by personality and behavioral changes, deficits on executive tasks, apathy or inertia and disinhibition (Rascovsky, Hodges et al. 2011, Rascovsky, Grossman 2013). On the other hand, in PPAs, language and speech are predominantly affected (Gorno- Tempini, Hillis et al. 2011). The syndrome is subdivided into non-fluent variant primary progressive aphasia (nfvPPA) characterized by apraxia (Rosen, Allison et al. 2006), and semantic variant primary progressive aphasia (svPPA) predominantly manifesting in behavioral symptoms as well as difficulties in understanding words and naming objects (Gorno-Tempini, Hillis et al. 2011, Rosen, Allison et al. 2006). In addition, a logopenic variant (l-vPPA) has been identified, but the neuropathology of the disease is more closely associated with AD than FTLD and, hence, it is not included in the FTLD syndromes (Gorno-Tempini, Hillis et al. 2011). Moreover, at least 15 % of patients develop a motor-neuron disease, the most common of which being amyotrophic lateral sclerosis (ALS) (Burrell, Kiernan et al.

2011). It is noteworthy, that the phenotypes often converge during the disease progression (Mackenzie, Neumann 2016).

FTLD is heterogenous also in neuropathological terms. It is divided into tau positive and tau negative forms (Mackenzie, Neumann et al. 2010). The most common FTLD neuropathology is the tau-negative subtype, FTLD-TDP-43, accounting for approximately 50% of cases (Mackenzie, Neumann 2016), whereas the tau positive forms can be observed in approximately 35% of FTLD cases. In the tau-positive subtype, tau protein is hyperphosphorylated leading to the formation of NFTs as in AD (Cairns, Bigio et al. 2007).

10

Approximately 25-50% of FTLD cases have a positive autosomal dominant family history of the disease, e.g. involving mutations in microtubule associated protein tau (MAPT), progranulin (GRN) and hexanucleotide repeat expansion in chromosome 9 open reading frame 72 (C9ORF72) (Rademakers, Neumann et al. 2012).

2.2.2 Vascular dementia

VaD is applied as a general term to describe the cognitive disorders caused by vascular brain damage. The severity can range from vascular cognitive impairment (VCI), which is equivalent to MCI, up to full-blown dementia. VaD is estimated to account for approximately 20% of memory disorders, and co-existence with other neurodegenerative diseases, especially with AD, is common. It has been proposed that the combination of AD and vascular pathology may gradually become the most common cause of dementia in the future (Jellinger, Attems 2015, Rahimi, Kovacs 2014). As in AD, the incidence of VaD increases exponentially after 65 years of age (Jellinger 2013). In the age group of 65 to 69 years, the prevalence of VaD is higher in men than in women (0.5% and 0.3%, respectively), but after 85 years of age, women outnumber men (5.8% and 3.6%, respectively) (Jellinger 2013).

The most common pathology behind VaD is cerebral small vessel disease (approximately 80% of all cases), which leads to various subcortical changes; white matter changes, lacunar infarcs, enlarged perivascular spaces and atrophy. Cerebral amyloid angiopathy (CAA) is a hermorrhagic form of cerebral small vessel disease. (Attems, Jellinger et al. 2011). Other vascular causes are infarcts due to large or middle-size-artery atherosclerosis (previously called multi-infarct dementia), and strategic infarcts (Thal, Grinberg et al. 2012, Kalaria 2017).

Considering the etiology of VaD, it is not surprising that several common vascular risk factors, such as hypertension, diabetes and obesity, as well as age, are independent risk factors also for VaD as is the case for the other dementias (Gorelick, Scuteri et al. 2011).

Furthermore, several genetic risk factors for cerebral small vessel disease have been identified, with the most important being mutations in the Notch homolog 3 (NOTCH3) gene which is implicated in cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) (Chabriat, Joutel et al. 2009, Low, Junna et al. 2007).

2.2.3 Dementia with Lewy bodies

Dementia with Lewy bodies is the second most common cause of neurodegenerative dementia after AD in the elderly. It is responsible for approximately 10 % of all cases, but as many as half of these are actually combinations of DLB and AD pathology. The onset of DLB typically occurs between 50 to 80 years. (McKeith, Mintzer et al. 2004).

Fluctuating cognition with pronounced variations in attention and alertness are typical symptoms of DLB. Detailed and well-formed visual hallucinations occur in up to 80% of the patients; these are common already in the early stages of the disease and appear to persist also later in the course of the illness. On the contrary, rapid eye movement (REM) sleep behavior disorder associated with the absence of normal REM atonia, and hence allowing the individuals to “act out their dreams”, often precedes other symptoms by many years but may become less dominating with disease progression (McKeith, Boeve et al. 2017, Boeve, Silber et al. 2004). Moreover, as many as 85% of DLB patients eventually exhibit parkinsonianism including rigidity and problems with gait (McKeith, Boeve et al. 2017). However, in DLB the impairments in cognition appear before the motor symptoms, where as in Parkinson’s disease (PD), cognitive problems are encountered only years after the appearance of physical symptoms (McKeith, Mintzer et al. 2004, Donaghy, McKeith 2014). In addition to the core criteria mentioned earlier, patients with DLB are typically very sensitive to antipsychotic drugs and they may experience repeated falls and suffer from severe autonomic dysfunction (McKeith, Boeve et al. 2017).

11

The pathological features in the brain in DLB are the so-called Lewy bodies and Lewy neurites consisting of aggregated α-synuclein. These alterations can also be found in the brain of subjects with PD (McKeith, Mintzer et al. 2004). Most patients with DLB also have Aβ- plaques and NFTs typical for AD, and mixed dementia AD+DLB may be more common than pure DLB (Schneider, Arvanitakis et al. 2007, Forsberg, Almkvist et al. 2010). This modifies the clinical presentation of the disease (McKeith, Mintzer et al. 2004, Ballard, Gauthier et al.

2011, Merdes, Hansen et al. 2003).

2.2.4 Creutzfeldt-Jakob disease

Prion diseases are rapidly progressing neurodegenerative diseases that are caused by the misfolding of a host-encoded cellular prion protein, PrP^C, in the brain (Imran, Mahmood 2011). Creutzfeldt-Jakob disease (CJD) is the most common of the prion diseases.

The exceptionally rapid progression of cognitive impairment and the early appearance of focal neurological symptoms differentiate CJD from the other dementing diseases. The first symptoms of CJD classically include a rapidly progressive multifocal dementia and psychiatric symptoms. These are followed by myoclonus; dyskinesia is also common. In addition, some patients experience cortical blindness and visual hallucinations. (Imran, Mahmood 2011, Wadsworth, Collinge 2007).

The neuropathological hallmarks of the disease include the presence of spongiform changes and PrP^Sc reactivity in the brain tissue (Imran, Mahmood 2011, Wadsworth, Collinge 2007). In approximately 5% of the sporadic CJD cases, also deposits of amyloid plaques can be detected (Imran, Mahmood 2011, Johnson 2005). In addition, the concentration of tau protein in the CSF of CJD patients is exceptionally high due to the widespread neuronal damage. Moreover, the concentration of protein 14-3-3 depicting rapid neuronal degeneration is increased (Wadsworth, Collinge 2007).

2.2.5 Normal pressure hydrocephalus

Normal pressure hydrocephalus (NPH) is a disease classically characterized by gait disturbances, progressive cognitive decline and urinary incontinence. It has been claimed to be responsible for up to 6% of dementia cases (Savolainen, Paljärvi et al. 1999, Graff-Radford 2007), and its prevalence increases with age (Brean, Eide 2008).

The pathological hallmarks of NPH are enlarged ventricles and normal or slightly elevated intracranial pressure (Pyykkö, Lumela et al. 2014, Leinonen, Koivisto et al. 2012). In idiopathic NPH, concomitant AD and VaD-type pathology, such as the accumulation of Aβ- plaques and tau protein, and changes in the concentrations of these proteins in the CSF, are often detected (Pyykkö, Lumela et al. 2014, Leinonen, Menon et al. 2011). However, the underlying mechanisms underpinning the production of Aβ appears to be different (Laiterä, Sarajärvi et al. 2014). Neuroimaging can reveal the atrophy of hippocampus and the surrounding areas, which is common in NPH, but can also refer to the possibility of concomitant AD (Graff-Radford 2007). Apparently, patients with concomitant AD-type pathology are likely to develop dementia later (Leinonen, Koivisto et al. 2012) .

2.3 MILD COGNITIVE IMPAIRMENT

Over time, numerous terms have been used to describe the transitional phase between cognitively normal aging and early dementia. Today, the most widely used term for this condition is mild cognitive impairment (MCI). In MCI, there is a deficit in one or some cognitive domains but daily functional abilities are spared, and the person does not meet the criteria for dementia (Petersen 2004).

12

MCI can be divided into multiple subgroups based on their clinical picture. They also have different kinds of etiologies, and hence also different outcomes (Figure 5). In amnestic MCI (a-MCI), memory is the only cognitive domain that is affected; these persons are likely to convert to AD. In contrast, if multiple cognitive domains are affected, the condition is called multidomain MCI (md-MCI). This group can be further divided into amnestic and non- amnestic subgroups. Individuals in the former subgroup are likely to have AD or VaD, or a combination of these diseases. Persons in the latter subgroup probably have DLB or VaD, or both. Finally, if there is a deficit in one cognitive domain excluding memory, the condition is called single non-memory MCI (Petersen 2004, Petersen, Roberts et al. 2009).

Figure 5. The classification of MCI subtypes, and their hypothesized etiology. Modified from Winblad et al 2004 and Petersen et al 2009.

The prevalence of MCI appears to be approximately four times that of dementia. The progression rate of MCI into dementia has varied across studies being approximately 10-15%

per year (Eshkoor, Hamid et al. 2015). In comparison with the general population, in the study of Petersen et al, only 1-2% converted from normal status to dementia in a single year.

When the MCI patients were followed further, within six years as many as 80% had developed dementia (Petersen 2004).

It should be noted that neurodegenerative diseases are not the only possible etiology behind MCI. It can also be caused by psychiatric disorders and metabolic disturbances (Winblad, Palmer et al. 2004) as well as excessive alcohol consumption (Anttila, Helkala et al. 2004). The diagnosis of MCI is based on the same methods as in dementia: clinical examination, neuropsychological testing, laboratory tests, imaging and assessment of different kinds of biomarkers (Eshkoor, Hamid et al. 2015). Brain imaging may reveal hippocampal and temporal lobe atrophy (Eshkoor, Hamid et al. 2015, DeCarli, Miller et al.

2001). Furthermore, in MCI later progressing into AD, the concentration of Aβ in the CSF may already be decreased and tau concentrations increased, but not as extensively as in AD (Herukka, Hallikainen et al. 2005, Hansson, Zetterberg et al. 2006, Olsson, Lautner et al. 2016).