DISSERTATIONS | BUSHRA IMTIAZ | HORMONE THERAPY AND THE RISK OF DEMENTIA,COGNITIVE... | No 397
uef.fi
PUBLICATIONS OF
THE UNIVERSITY OF EASTERN FINLAND Dissertations in Health Sciences
ISBN 978-952-61-2402-5 ISSN 1798-5706
Dissertations in Health Sciences
THE UNIVERSITY OF EASTERN FINLAND
BUSHRA IMTIAZ
HORMONE THERAPY AND THE RISK OF DEMENTIA, COGNITIVE DECLINE AND ALZHEIMER’S DISEASE
Depletion of estrogen and progesterone at menopause may predispose to cognitive
decline and Alzheimer’s disease (AD).
Hormone therapy (HT) has been suggested to prevent or delay this. The findings from previous studies have been inconsistent. AD-
HT association is a complex scenario and is subjected to various genetic and lifestyle factors. This thesis explored the direction of association between HT, AD, and cognition in two nation-wide case-control studies and two
longitudinal cohort studies.
BUSHRA IMTIAZ
Hormone Therapy and the Risk of Dementia,
Cognitive Decline and Alzheimer’s disease
BUSHRA IMTIAZ
Hormone Therapy and the Risk of Dementia, Cognitive Decline and Alzheimer’s disease
To be presented by permission of the Faculty of Health Sciences, University of Eastern Finland for public examination in Auditorium CA100, Canthia building of the University of Eastern Finland,
Kuopio, on Friday, January 20th 2017, at 12 noon
Publications of the University of Eastern Finland Dissertations in Health Sciences
Number 397
Institute of Clinical Medicine-Neurology School of Medicine, Faculty of Health Sciences
University of Eastern Finland Kuopio
2017
Series Editors:
Professor Tomi Laitinen, M.D., Ph.D.
Institute of Clinical Medicine, Clinical Physiology and Nuclear Medicine Faculty of Health Sciences
Professor Hannele Turunen, 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): 978-952-61-2402-5
ISBN (pdf): 978-952-61-2403-2 ISSN (print): 1798-5706
ISSN (pdf): 1798-5714 ISSN-L: 1798-5706
Author’s address: Department of Neurology, Institute of Clinical Medicine, School of Medicine University of Eastern Finland
KUOPIO FINLAND
Supervisors: Professor Hilkka Soininen, M.D., Ph.D.
Department of Neurology, Institute of Clinical Medicine, School of Medicine University of Eastern Finland
KUOPIO FINLAND
Professor Miia Kivipelto, M.D., Ph.D.
Department of Neurology, Institute of Clinical Medicine, School of Medicine University of Eastern Finland
KUOPIO FINLAND
Center for Alzheimer Research, Division for Neurogeriatrics Department of Neurobiology, Care Sciences and Society Karolinska Institutet
STOCKHOLM SWEDEN
Assistant Professor Anna-Maija Tolppanen, PhD.
Faculty of Health Sciences, Department of Pharmacy University of Eastern Finland
KUOPIO FINLAND
Reviewers: Professor Pirkko Härkönen, M.D., Ph.D.
Institute of Biomedicine, Cell Biology and Anatomy University of Turku
TURKU FINLAND
Professor Kaisu Pitkälä, M.D., Ph.D.
Department of General Practice and Primary Health Care University of Helsinki
HELSINKI FINLAND
Opponent: Docent Kati Juva, M.D., Ph.D.
Division on Psychiatry Helsinki University Hospital HELSINKI
FINLAND
Imtiaz Bushra
Hormone Therapy and the Risk of Dementia, Cognitive Decline and Alzheimer’s disease University of Eastern Finland, Faculty of Health Sciences, 2017
Publications of the University of Eastern Finland. Dissertations in Health Sciences 397. 2017. 88 p.
ISBN (print): 978-952-61-2402-5 ISBN (pdf): 978-952-61-2403-2 ISSN (print): 1798-5706 ISSN (pdf): 1798-5714 ISSN-L: 1798-5706
ABSTRACT
The depletion of female sex steroid hormones occurring at menopause (natural/induced) might expose women to an increased risk of dementia, cognitive decline, and Alzheimer’s disease (AD). The availability of sex steroid hormone receptors (estrogen and progesterone) in wide areas of the brain may represent a biologically plausible mechanism for the higher AD risk in females after they pass through menopause. Findings from previous observational studies and clinical trials of the association between AD and hormone therapy (HT) use are ambiguous.
This thesis comprises four research articles, two of which (1 & 4) are register based nationwide case-control studies while the other two (2 & 3) are longitudinal cohort studies.
Register based studies are based on data from the Medicine and Alzheimer’s disease (MEDALZ) cohort with clinically verified AD diagnosis from 1999-2005 for study 1 (n of matched case control pairs = 19,043) and from 2005-2011 for study 4 (n of AD cases = 46,117 and n of controls= 184,463). Kuopio Osteoporosis Risk factors and Prevention (OSTPRE) cohort (n = 8195) and Cardiovascular Risk Factors, Aging and Dementia (CAIDE) cohort (n = 731) comprise the study population for studies 2 and 3 respectively. The mean follow-up time for study 1 was from 1986-2005; in study 4, it was from 1995-2011. The mean follow-up time for study 2 was 20 years; for study 3, it was 8.3 years. The main outcome for studies 1, 2 and 4 was probable AD based on Diagnostic and Statistical Manual of Mental Disorders (DSM- IV) and National Institute of Neurologic and Communicative Disorders and Stroke and the Alzheimer’s disease and Related Disorders Association (NINCS-ADRDA) criteria; this was extracted from the Finnish special reimbursement register, while for study 3, the main outcome was the cognitive status measured at baseline (1998) and follow-up (2005). The use of HT was the main exposure in studies 2, 3 and 4 whereas gynecological surgeries (oophorectomy, hysterectomy, and hysterectomy with bilateral oophorectomy) were the main exposure in study 1. Data on surgical procedures was taken from the hospital discharge register while register-based data on HT use was collected from 1995 onwards from the prescription register for studies 2, 3, and 4 along with self-reported use of HT in studies 2 and 3.
The majority of women in all four studies were postmenopausal with respect to the use of HT and status of surgery. Postmenopausal surgical removal of ovaries or uterus was not a significant predictor of AD irrespective of HT use and indication for surgery. Moreover, the long term use of postmenopausal HT was protective against AD independent of surgical status (study 1). Overall postmenopausal HT use was not significantly related to AD risk (studies 2, 4) or cognitive decline (study 3) unless the use of estrogen HT exceeded over 10 years, in that case, estrogen HT was protective against AD (studies 2, 4).
The results from this thesis indicate that the association of gynecological surgeries and HT with the risk of AD and cognitive decline depend upon the time of surgery with respect to the onset of menopause and duration of HT use. The protective association between longer duration of HT use with AD indirectly favors the critical window and healthy cell theories, where the effect of HT use depends upon the health status of neurons at baseline. One
promising avenue for future AD studies would involve clarifying the association of short- term HT use around menopause with late-onset AD and the use of brain-specific selective estrogen receptor modulators (SERM) which spare peripheral estrogen receptors to avoid estrogen-related peripheral adverse effects.
National Library of Medicine Classification: WT 155, WM 220, WP 580, WP 522, WP 468, WP 530
Medical Subjects Headings: Dementia; Alzheimer’s disease; Menopause; Estrogen; Progesterone; Hormone therapy; Cognitive decline; Oophorectomy; Hysterectomy; Hysterectomy with bilateral oophorectomy
Imtiaz Bushra
Hormone Therapy and the Risk of Dementia, Cognitive Decline and Alzheimer’s disease Itä-Suomen yliopisto, terveystieteiden tiedekunta, 2017
Publications of the University of Eastern Finland. Dissertations in Health Sciences 397. 2017. 88 s.
ISBN (print): 978-952-61-2402-5 ISBN (pdf): 978-952-61-2403-2 ISSN (print): 1798-5706 ISSN (pdf): 1798-5714 ISSN-L: 1798-5706
TIIVISTELMÄ:
Naisten estrogeeni- ja progesteronitasot laskevat vaihdevuosien aikana. Vaihdevuodet ovat useimmin luonnolliset, mutta voivat johtua myös kohdun ja munasarjojen poistosta. Tämä hormonaalinen muutos yhdessä aivoissa sijaitsevien estrogeeni- ja progesteronireseptoreiden kanssa voivat selittää naisten miehiä suuremman alttiuden tiedollisten toimintojen alenemiselle tai muistisairauksille, kuten dementian yleisin muoto Alzheimerin tauti. Tätä hypoteesia tukee myös se, että aivojen useilla alueilla on estrogeeni- ja progesteronireseptoreja. Aiempien hormonihoitojen ja Alzheimerin taudin yhteyttä selvittäneiden havainnoivien tutkimusten ja satunnaistettujen kontrolloitujen kokeiden löydökset eivät kuitenkaan ole olleet yksiselitteisiä.
Tämä väitöskirja koostuu neljästä osatyöstä, joista kaksi (1 & 4) pohjautuu valtakunnallisiin rekisteripohjaisiin tapaus-verrokki tutkimuksiin ja kaksi (2 & 3) pitkittäisiin kohorttitutkimuksiin. Osatyöt 1 ja 4 perustuvat Medicine and Alzheimer disease (MEDALZ)- tutkimuksiin, joissa osatyössä 1 olivat mukana kaikki 31.12.2005 elossa olleet suomalaiset, joilla oli kliinisesti varmennettu todennäköisen Alzheimerin taudin diagnoosi (19043 tapausta ja 19043 verrokkia) ja osatyössä 4 kaikki ne, jotka saivat saman diagnoosin vuosina 2005–2011 (46117 tapausta ja 184463 verrokkia). Osatyössä 2 käytettiin aineistoa Kuopio Osteoporosis Risk Factors and Prevention (OSTPRE)-tutkimuksesta (n=8195) ja osatyössä 3 Cardiovascular Risk Factors, Aging, and Dementia (CAIDE)-tutkimuksesta (n= 731).
Osatöissä 1, 2 ja 4 päätetapahtumana oli todennäköinen Alzheimerin taudin diagnoosi perustuen Diagnostic and Statistical Manual of Mental Disorders- (DSM-IV) ja National Institute of Neurologic and Communicative Disorders and Stroke and the Alzheimer’s disease and Related Disorders Association- (NINCS-ADRDA) kriteeristöihin. Diagnoosit poimittiin erityiskorvausoikeusrekisteristä. Osatyössä 3 tutkittiin tiedollisten toimintojen alenemista eri osa-alueilla kahdeksan vuoden seurannan aikana. Osatyössä 1 (seuranta-aika 1995–2005) tutkittiin kohdun ja/tai munasarjojen poistoa Alzheimerin taudin riskitekijänä ja sitä, selittyykö yhteys hormoniihoidon käytöllä. Osatöissä 2-4 tutkittiin hormonihoidon yhteyttä tiedollisten toimintojen alenemaan ja Alzheimerin tautiin. Tieto kohdun ja/tai munasarjojen poistosta poimittiin hoitoilmoitusrekisteristä ja hormonihoidon käyttö osatöissä 2-4 reseptirekisteristä. Lisäksi tieto hormonihoidon käytöstä kerättiin kyselylomakkein osatöissä 2 ja 3. Kaikissa neljässä osatyössä naiset olivat pääosin vaihdevuosi-iän ohittaneita. Osatyössä 2 seuranta-aika oli 20 vuotta, osatyössä 3 keskimäärin 8.3 vuotta ja osatyössä 4 1995–2011.
Kohdun ja/tai munasarjojen poisto vaihdevuosi-iän jälkeen ei ollut merkittävä Alzheimerin taudin riskitekijä. Tulokset olivat samanlaiset riippumatta hormonihoidosta ja kirurgian indikaatioista. Osatyössä 1 hormonihoitoa pitkään käyttäneillä oli pienempi Alzheimerin taudin riski kuin niillä, jotka eivät olleet käyttäneet hormonihoitoa, riippumatta siitä oliko heille tehty kohdun ja/tai munasarjojen poisto. Hormonihoitoa käyttäneiden naisten Alzheimerin taudin riski (osatyöt 2 ja 4) sekä tiedollisten toimintojen aleneminen (osatyö 3) olivat yhtä suuria kuin niillä, jotka eivät käyttäneet hormonihoitoa. Osatöissä 2 ja 4 havaittiin kuitenkin, että yli 10 vuotta hormonihoitoa käyttäneillä oli pienempi Alzheimerin taudin riski.
Tämän väitöskirjan tulosten perusteella kohdun ja/tai munasarjojen poiston ja hormonihoidon sekä tiedollisten toimintojen alenemisen ja muistisairauksien välinen yhteys riippuu hormonihoidon kestosta sekä siitä, onko toimenpiteet tehty vaihdevuosi-iän jälkeen.
Pitkäkestoisen hormonihoidon yhteys pienempään Alzheimerin taudin riskiin voi selittyä kriittisen aikaikkunan teorialla. Jatkossa olisikin tärkeä tutkia vaihdevuosiin ajoittuvaa lyhytkestoista hormonihoitoa sekä selektiivisten estrogeenireseptoriin vaikuttavien modulaatoreiden (SERM) kohdentamista aivoihin.
Luokitus: WT 155, WM 220, WP 580, WP 522, WP 468, WP 530
Yleinen Suomalainen asiasanasto: Dementia; Alzheimerin tauti; Menopaussi; Estrogeeni; Progesteroni;
Hormonihoito; kognitio; Munasarjojen poisto; Kohdunpoisto
To My Family
Acknowledgements
This Ph.D. thesis was completed in the Institute of Clinical Medicine, Department of Neurology, Faculty of Health Sciences, Kuopio, University of Eastern Finland under the Doctoral Program of Molecular Medicine.
Firstly, I will like to express my immense gratitude to Professor Hilkka Soininen, M.D., Ph.D., my principal supervisor and mentor, for believing in me and providing me with an opportunity to prove myself. Her clear vision, quick responses, support, and expert opinions were the fuel that powered me through these years. I want to thank from bottom of my heart my secondary supervisors Professor Miia Kivipelto M.D. Ph.D. and Associate Professor Anna-Maija Tolppanen Ph.D. for their expert guidance and invaluable contributions towards completion of my doctoral research. Anna-Maija Tolppanen, I cannot thank you enough for all the hard work and encouragement you have poured into my research to make it better in so many ways. Your to-the-point and timely responses have always left me in awe. I found a friend in you who was always there to help me whenever I needed it.
I sincerely thank all co-authors of my publications (in alphabetical order) Anna-Mari Heikkinen, Antti Tanskanen, Alina Solomon, Heidi Taipale, Heikki Kröger, Jari Tiihonen, Miia Tiihonen, Marjo Tuppurainen, Sirpa Hartikainen andToni Rikkonen for their useful insights, comments and critique. All of them are very competent and I have learnt a lot from them. I want to especially thank Associate Professor Alina Solomon M.D. Ph.D. for her deep concern, critical advice and brief but stimulating, meetings to improve my morale and knowledge alike.
My warmest thanks goes to the pre-examiners of this thesis, Professor Kaisu Pitkälä, M.D., Ph.D. and Professor Pirkko Härkkönen, M.D., Ph.D. for their in-depth critical comments and suggestions, which improved my thesis considerably. I express my gratitude to Docent Kati Juva M.D., Ph.D. for agreeing to act as my opponent at the public examination of this Ph.D.
thesis.
I extend my thanks to Esa Koivisto for all technical help, Mari Tikkanen for travel and grant- related advices, and Arja Afflekt for her expert opinions regarding official documentation towards completion of my Ph.D. work. My sincere thanks goes to Ewen MacDonald Ph.D.
for the excellent linguistic revision of the thesis.
I want to acknowledge all my colleagues and office mates for their support, encouragement, and the creation of such a friendly working environment. Thank you Anna Rosenberg for helping me with the Finnish language, random things, and for daily friendly chit-chat. Many thanks to Marjo Eskelinen for her advice whenever I needed it. I owe special thanks to my colleague and best friend Ruth Stephens for everything whether it was just gossip, work- related discussions, family and social gatherings, we really excelled in all together. Thank you Olli Jääskeläinen for coffee breaks to wake-up our work routines. I will miss you all a lot.
My last set of thanks is for my amazing family without whom I am nothing. My special thanks to my parents who stood by me through thick and thin. It’s not a matter of just four years of Ph.D., it’s a span of decades of their hard work to enable me to achieve my goals.
Their vision is my target and their pride is my aim. I salute my mother for being such a unique and inspiring role model in all spheres of life. Her encouragemet, support and understanding towards me is precious and cannot be described in words. I want to express my heartiest gratitude towards my motherly Aunt Shahida Nasreen and fatherly Uncle Tahir Ijaz, you are part of my soul and no picture of mine is complete without both of you. I extend my warmest thanks to my loving and caring siblings namely Nomita, Ali, Tehreem, Shahnawaz, Haider, Alina and my dear sister-in-law and friend Zahra, for their understanding, support and confidence in me, you guys rock and I love you all.
I cannot thank enough my dearest husband Ali Hussain for his support, love, understanding and respect towards me. I remember the day when I embarked on this Ph.D. journey leaving a three years old daughter and a three months old newborn son at home. It was not easy at all but Ali believed in me more than I believed in myself and encouraged me to pursue my career as he knew it was equally important to me. We did it together and now it seems as if four years flew in a flash. I am thankful and humbled for our lovely kids, Jannat and Hashim, for their sunshine, warmth, and unconditional love and for being the best children that any parent could ever hope for. I would also like to thank all of my Pakistani community in Kuopio for arranging interesting parties which enabled me to maintain my sanity throughout these years.
I am the most grateful to the Almighty for all His blessings and favors.
Lastly, I would like to thank all the funding sources during my Ph.D. work namely University of Eastern Finland, Doctoral Program in Molecular Medicine, and Finnish Cultural Foundation.
Kuopio, January 2017 Bushra Imtiaz
List of the original publications
This dissertation is based on the following original publications:
I Imtiaz B, Tuppurainen M, Tiihonen M, Kivipelto M, Soininen H, Hartikainen S, Tolppanen AM. Oophorectomy, hysterectomy, and risk of Alzheimer's disease: a nationwide Case-Control study.
J Alzheimers Dis. 2014; 42(2):575-81. Doi: 10.3233/JAD-140336.
II Imtiaz B, Tuppurainen M, Rikkonen T, Kivipelto M, Soininen H, Kröger H,
Tolppanen AM. Post-menopausal hormone therapy and risk of Alzheimer’s disease:
a prospective cohort study (Accepted in Neurology)
III Imtiaz B, Tolppanen AM, Solomon A, Soininen H, Kivipelto M. Estradiol and cognition in the cardiovascular risk factors, aging, and dementia (CAIDE) cohort study. J Alzheimers Dis. 2016 Dec 9. [Epub ahead of print]
IV Imtiaz B, Taipale H, Tanskanen A, Tiihonen M, Kivipelto M, Heikkinen AM, Tiihonen J, Soininen H, Hartikainen S, Tolppanen AM. Risk of Alzheimer’s disease among postmenopausal hormone therapy users in a nation-wide case-control study.
(Submitted for publication)
The publications were adapted with the permission of the copyright owners.
Contents
1 INTRODUCTION………...1 2 REVIEW OF THE LITERATURE………..2 2.1 Cognition………..2 2.2 Dementia………..2
2.2.1 Causes and symptoms of dementia……….3
2.3 Alzheimer's disease……….…4 2.3.1Diagnostic criteria for Alzheimer's disease………..………5 2.3.2 Biomarkers of Alzheimer's disease……….………..6 2.3.3 Midlife risk and protective factors of Alzheimer's disease ………..8 2.3.4Sex and gender based dimorphism in brain and Alzheimer's disease………….…10 2.4 Female sex steroid hormones………..…11 2.4.1 Estrogens and progesterone……….12 2.4.2 Mechanisms of neuroprotective effects of estrogen and progesterone……….12 2.4.3 Sex steroid hormones and aging brain ………..14 2.4.4 Types of commercially available hormone therapy and their implications………15 2.5 Menopause……….17 2.5.1 Types and stages of menopause………17 2.5.2 Short-term and long-term implications of menopause……….………..18 2.6 Hormone therapy and risk of Alzheimer's disease and dementia……….18 2.6.1 Clinical trials……….19 2.6.2 Observational studies………..25 2.6.3 Surgical menopause and hormone therapy and Alzheimer's disease………..31 3 AIMS OF THE STUDY………...34 4 SUBJECTS AND METHODS………..35 4.1 Medicine and Alzheimer's disease (MEDALZ) study (Study 1 and 4)………..35 4.1.1 Study population and design………...35 4.1.2 Exposure data………35 4.1.3 Outcome data……….37 4.1.4 Covariables………...…….37 4.2 Kuopio Osteoporosis Risk Factors and Prevention (OSTPRE) cohort (Study 2)………..38 4.2.1 Study population and design………..38 4.2.2 Exposure data………40 4.2.3 Outcome data……….40 4.2.4 Covariables………....40 4.3 Cardiovascular Risk Factors, Aging and Dementia (CAIDE) cohort (Study 3)………...41 4.3.1 Study population and design……….….41 4.3.2 Exposure data………42 4.3.3 Outcome data……….…42 4.3.4 Covariables………42 4.4 Statistical analyses……….……43 4.4.1 Study 1………43 4.4.2 Study 2………43 4.4.3 Study 3………43
4.4.4 Study 4………44 5 RESULTS………...45 5.1 Baseline characteristics of study population……….45 5.2 Oophorectomy, hysterectomy, radical hysterectomy and Alzheimer's disease …..…...48 5.3 Hormone therapy use and Alzheimer's disease ……….……….49 5.4 Hormone therapy and cognitive decline………51 6 DISCUSSION……….53 6.1 Oophorectomy, hysterectomy, and risk of Alzheimer's disease (Study 1)………53 6.2 Postmenopausal hormone therapy use and risk of Alzheimer's disease and dementia (Studies 1, 2, 4)……….54 6.2.1 Use of hormone therapy in relation to oophorectomy and hysterectomy (Study 1)…….………...54 6.2.2 Postmenopausal hormone therapy and risk of Alzheimer's disease and dementia (Studies 2, 4)……….55 6.3 Use of hormone therapy and cognitive decline (Study 3)………56 6.4 Methodological considerations (all studies)………..57 7 CONCLUSIONS………60 8 FUTURE PERSPECTIVES………...61 9 REFERENCES……….63 ORIGINAL PUBLICATIONS (I-IV)
Abbreviations
Aβ amyloid beta protein AD Alzheimer’s disease ADRDA Alzheimer’s disease and
related disorders association APP amyloid precursor protein APOE apolipoprotein E (gene) ATC Anatomical Therapeutic
Chemical BMI body mass index Bcl-2 B cell lymphoma 2
CAIDE Cardiovascular Risk Factors, Aging and Dementia
CEE conjugated equine estrogens CERAD Consortium to Establish a
Registry for Alzheimer’s disease
CI confidence interval CSF cerebrospinal fluid
CT computerized tomography DSM-IV Diagnostic and Statistical
Manual of Mental Disorders 4th edition
DLB Dementia with Lewy bodies ELITE Early versus Late Intervention
Trial with Estradiol FINMONICA Finnish Multinational
Monitoring of Trends and Determinants in
Cardiovascular disease FTLD frontotemporal lobar
degeneration G03C estrogen G03D progesterone
G03F estrogen and progesterone in combination
G03X other sex hormones and modulators of genital system HR hazards ratio
HRQOL Health-related quality of life HT hormone therapy
ICEE index of cumulative estrogen exposure
ICD International Classification of Diseases
IWG International Working Group KEEPS-cog Kronos Early Estrogen
Prevention Study-cognitive and affective study
MEDALZ Medicine and Alzheimer’s disease study
MCI mild cognitive impairment MMSE mini-mental scale
examination
MPA medroxy progesterone acetate MRI magnetic resonance imaging 3MS modified mini-mental scale NA not available
NIA-AA National Institute of Aging and Alzheimer’s Association NINCS-ADRDA National Institute of
Neurologic and
Communicative Disorders and Stroke and the Alzheimer’s disease and Related Disorders Association NOMESCO Nordic Medico-Statistical
Committee’s Classification of Surgical Procedures
NSAIDS Non-steroidal anti- inflammatory drugs
OSTPRE Osteoporosis Risk Factor and Prevention
OR odds ratio
PDD Parkinson’s disease dementia PET positron emission
tomography
PUFA poly-unsaturated fatty acids RCT randomized controlled trial SERM selective estrogen receptor
modulators
TICS-m Telephone interview of cognitive status-modified TSEC tissue specific estrogen
complex
USA United States of America VaD vascular dementia VMS vasomotor symptoms WHIMS Women’s Health Initiative
Memory Study
1 Introduction
Dementia is a syndrome characterized by progressive impairment in memory and other cognitive abilities, ultimately interfering with daily life activities. Dementia is attributable to Alzheimer’s disease (AD) in 60-80% of cases and in view of its health care and care-giving costs, many recent interventions have been aimed towards prevention, delay in onset, and to slow down progression of AD (Alzheimer's Association 2016). The world-wide prevalence of dementia in 2015 was 46.8 million, a value which is projected to double every 20 years leading to an estimated total of 131.5 million AD patients in 2050. Another way of assessing this change is that by 2030, AD will be a trillion dollar disease compared to the billion dollar disease it is at the moment (Alzheimer's disease International 2015).
Recent guidelines for AD diagnosis emphasize the importance of the identification of the pre-clinical stage of dementia based on clinical presentation and biomarker findings (cerebrospinal fluid (CSF), neuroimaging) in order to target preventive strategies for those at risk of dementia, since it is thought that they will benefit most from effective therapeutics (Dubois et al. 2010, Dubois et al. 2014). Research on the pathogenesis of AD has identified multiple modifiable midlife risk factors, thus providing a window of opportunity for prevention (Barnes and Lee 2011).
In the United States (US), two thirds of all AD cases are women (Hebert et al., 2013). The higher risk of AD in women might be due to their longer life span (Seshadri et al., 1997) or to the depletion of sex-steroid hormones at menopause (Vest and Pike 2013). This thesis focuses on the interaction between postmenopausal hormone therapy (HT) use and AD, and also investigates the cognitive decline in different population based studies from Finland.
Neuroprotective effects of estrogen are well-established in basic science and animal studies but findings from clinical trials and observational studies have been inconsistent (Brinton, 2008a, Zandi et al., 2002a, Bove et al., 2014, Shumaker et al., 2013, Espeland et al., 2004, Marinho et al., 2008). The presence of estrogen receptors in those areas of brain which are involved in AD pathology, provide support for a biologically plausible mechanism to account for the potential beneficial effects of HT to prevent this debilitating illness.
The present thesis comprises two case-control and two cohort studies and is intended to evaluate the effect of postmenopausal HT on late-life cognitive decline and AD. One study focuses on the association between AD and gynecological surgeries (oophorectomy, hysterectomy, and hysterectomy with bilateral oophorectomy). The studies in this thesis have evaluated various lifestyle, socioeconomic, and demographic variables in order to account for potential bias and confounding in the AD-HT relationship.
2 Review of the literature
2.1 COGNITION
Cognition comprises a set of mental abilities needed to perform a wide range of functions i.e. from the simplest to the most difficult task to be tackled. Cognition has been categorized into different domains depending upon the brain areas involved in their execution.
Cognitive domains can be subdivided into perception, attention, memory, motor skills, language, visuospatial abilities, and executive functions (Michelon Pascale, 2006). Aging itself is associated with a subtle cognitive decline but the cognitive decline associated with normal aging is different from the cognitive changes occurring in the preclinical stages of AD. Identification of AD-specific cognitive domains and their assessment is used in diagnostics of AD when the disease is in its initial stages. A decline in episodic memory of hippocampal type, which is not corrected by cueing recall, has been added to the diagnostic criteria in recent guidelines (Dubois et al., 2007). Other cognitive domains affected by aging are verbal memory (Marquis et al., 2002), working memory (Small et al., 1999), attention (Stankov, 1988), and visual perception (Koss et al., 1991). The presence of estrogen receptors in hippocampus, frontal lobes, and basal forebrain where most of these cognitive domains are controlled, represents a plausible connection between sex steroid hormones and cognitive process.
2.2 DEMENTIA
Dementia is an umbrella term characterized mainly by loss of memory and other mental abilities, ultimately leading to impairments in activities of daily living. Its incidence and health- and care-giving related costs are expected to increase at an alarming rate in the coming decades (Wimo et al., 2013).
Life expectancy has considerably increased during the last century, thus increasing the prevalence and incidence of chronic illnesses such as dementia. The regional prevalence of dementia ranges from 4.6% in Central Europe to 8.7% in North Africa and the Middle East.
The expected prevalence of dementia has been predicted to be much lower in high income countries compared to low and middle income countries where the prevalence is anticipated to increase from 58% to 68% between 2015 and 2050 (Alzheimer's disease International 2015). Dementia not only affects the diseased person but also the whole family in terms of care-giving and financial management. From a wider perspective, it affects entire societies and countries in terms of arranging health care services and increasing the burden of disease due to disabilities (Wimo et al., 2010, Wimo et al., 2013).
Currently, the worldwide costs of dementia are estimated to be 818 billion United States dollars, and dementia is expected to become a trillion dollar disease by 2030 (Wimo et al., 2016). Considering the continuous increase in the ageing population, dementia has been declared a global health priority by the United States, the Group of 7, and World Health Organization who have issued a common agenda of preventing and treating this debilitating disease through increased funding for its research and raising awareness through various public health and societal platforms at community levels (Wortmann 2012, Wimo et al., 2016).
2.2.1 Causes and symptoms of dementia
The most important symptoms of dementia include memory loss, difficulty in learning new information, language problems, and impairment in other cognitive domains namely planning, judgment, reasoning, organizing, and execution of complex tasks. Considering the complexity of the structure and function of the human brain, a direct diagnosis of dementia is always difficult and exclusion of other differential diagnoses is required in order to reach a definite diagnosis, in contrast to other chronic illnesses e.g. cardiovascular disease and cancer where direct diagnosis is relatively straightforward.
Figure 1: Etiologies of dementia with their subtypes (Alzheimer's Association 2016)
The different etiologies of dementia are summarized in Figure 1. AD is the most important type of dementia accounting for 60-80% of cases, and it is characterized by impaired memory as its primary symptom, with difficulties in memorizing newly learnt information (Wilson et al., 2012). The second most common (15-20%) type of dementia is vascular dementia (VaD) with initial symptoms of impaired judgement, planning, decision making, as well as motor functions being affected due to blockage or rupture of brain vessels leading to infarcts and hemorrhage. VaD is a heterogeneous condition with several subtypes including large vessel VaD (multi-infarct dementia, strategic infarct dementia); ischemic hypoperfusive VaD (cortical or subcortical); small-vessel VaD; and hemorrhagic VaD (Roman et al., 2002). Other common types of dementia are dementia with Lewy bodies (DLB) and Frontotemporal Lobar Dementia (FTLD) dementia, and Parkinson’s disease dementia (PDD). DLB and FTLD account for 10-15 % and 10% of disease burden respectively. DLB and Parkinson’s disease dementia (PDD) are characterized by the
Dementia [Alzheimer’s association, 2016]
Alzheimer’s dementia (AD) 60–80 %
Vascular dementia (VaD) 15–20 %
Dementia with Lewy bodies (DLB) 10-15 %
Parkinson’s disease dementia (PDD)
Frontotemporal Lobar Degeneration (FTLD) 10 %
Normal pressure hydrocephalus (5 %)
Almost half have only AD pathology Others have mixed pathology
Multi-infarct dementia Strategic infarct dementia Ischemic-hypoperfusive dementia Haemorrhagic vascular dementia
Others
Corticobasal degeneration Progressive supranuclear palsy
Pick’s disease Behavioral variant FTLD Creutzfeldt Jakob disease
Huntington’s disease Wernicke-Korsakoff Syndrome
presence of the abnormal accumulation of alpha synuclein protein (Lewy bodies) in neurons (Spillantini et al., 1998, McKeith 2002). FTLD presents with spared memory in the early stages (unlike AD), but instead with marked changes in personality and behavior (Rabinovici and Miller 2010). There are also other causes of dementia, e.g. dementia due to normal pressure hydrocephalus; Creutzfeldt Jakob disease related dementia due to a misfolded protein (prion) in brain; Huntington’s disease caused by a defective gene on chromosome 4; and Wernicke-Korsakoff syndrome, a chronic memory disorder caused by severe deficiency of thiamine (vitamin B-1) due to nutritional deficiency and alcoholism (Geldmacher and Whitehouse 1996, Zubaran et al., 1997). Mixed dementia is the type of dementia where more than a single cause of dementia is encountered. Around half of AD dementias have a mixed pathology e.g. AD+VaD, AD+DLB, AD+DLB+VaD; these are more frequent among the oldest old i.e. >85 years (Jellinger and Attems 2007).
2.3 ALZHEIMER’S DISEASE
AD is a neurodegenerative disorder which is not a part of normal aging. Its symptoms develop gradually and worsen over time, resulting in impairment of activities of daily living and ultimately death. In the US alone, 5.4 million people of all ages had been diagnosed with AD in 2016, 5.2 million of them were aged 65 years and older and this number will rise as the US older population (>65 years) is expected to increase from 14% in 2012 to 22% in 2050 (Alzheimer's Association 2016).
Histopathologically, AD is characterized by extracellular deposition of amyloid protein (Aβ) plaques and intracellular tau protein tangles. Previous studies have clarified the association between AD and neurofibrillary tangles (Bennett et al., 2004) and brain amyloidosis and AD (Mormino et al., 2009). Moreover, brain amyloid deposition has been linked with decreased brain volume (Oh et al., 2011, Fjell et al., 2010, Dickerson et al., 2009), disruption of functional networks (Sperling et al. 2009, Sheline et al. 2010, Hedden et al.
2009), and an increased risk of cognitive decline and progression to dementia in the future (Fagan et al., 2007, Li et al., 2007, Resnick et al., 2010) in cognitively normal older people.
These findings emphasize the role of brain amyloid and tau pathology in AD diagnosis. AD related pathology primarily involves brain areas involved with memory and it then spreads to other brain regions, thus evoking varying symptoms with different severities.
AD can be sporadic or late onset among >65 year olds as well as familial or early onset among younger people i.e. <65 years of age. Investigations into the familial form of AD, also known as monogenic AD, has helped in understanding the possible mechanisms initiating this illness. Familial AD is caused by autosomal dominant inherited gene mutations coding for amyloid precursor protein, presenilin 1 and presenilin 2. This kind of autosomal AD accounts for only about 1% of all AD cases. The increased risk of AD in subjects carrying mutations in these genes is proposed to be due to increased accumulation of amyloid proteins (Amemori et al., 2015). Sporadic AD develops possibly due to decreased clearance of amyloid from brain, either by carrying the Ɛ4 allele of apolipoprotein E (APOE), or being genetically predisposed due to a family history of AD, or due to still unknown mechanisms (Padayachee et al., 2016).
2.3.1 Diagnostic criteria for AD
The first AD case was diagnosed in 1906, but it was not until 1984 when the first diagnostic criteria for AD were devised by the National Institute of Neurological and Communicative Disorders and the Alzheimer’s disease and Related Disorders Association (NINCS- ADRDA) workgroup; these have now been used successfully for over 27 years. According to the NINCS-ADRDA criteria, AD could only have an amnestic presentation; it was considered to be a clinical and pathological entity, whose diagnosis was only probable during the subject’s lifetime; a definite diagnosis could be ascertained only after combining amnestic presentation during the patient’s life with AD-related neuropathology at autopsy.
No intermediate stage of memory loss was defined and information on biomarkers was lacking at that time as well as nothing being known about genetic forms. Furthermore non- amnestic presentations of AD were not acknowledged to any significant extent (McKhann et al., 1984).
Dementia research has recently viewed new horizons, i.e. AD pathology has been observed in the absence of clinical symptoms (Price and Morris 1999) along with non-amnestic clinical presentation of AD with prominent language and visuospatial abnormalities (Tang-Wai et al., 2004, Rabinovici et al., 2008, Alladi et al., 2007). The presence of AD pathology before the onset of clinical symptoms led to the assumption that there must be a long asymptomatic stage between the first brain lesion and the first clinical symptom, thus raising the importance of identifying this intermediate stage (Dubois et al., 2007, Dubois et al., 2010).
Based on recent developments, National Institute of Aging and Alzheimer’s Association (NIA-AA) sponsored a revision of the NINCS-ADRDA criteria (Jack et al., 2011, Sperling et al., 2011, Albert et al., 2011, McKhann G et al., 1984, McKhann et al., 2011). Moreover, an International Working Group (IWG) revised NINCS-ADRDA criteria separately and produced their recommendations for AD diagnosis [Dubois et al., 2014; Dubois et al., 2007;
Dubois et al., 2010].
The concept of dementia has been changed from a clinical-pathological entity (NINCS- ADRDA) to a dual clinical-biological entity in the IWG criteria and to a pathophysiological and clinical entity by NIA-AA. AD is now characterized as a spectrum of disease by NIA- AA, which has different stages inherent in its pathology; pre-clinical AD, mild cognitive impairment (MCI), and AD-dementia. Pre-clinical AD defined by NIA-AA is only intended for research purposes at the moment. Biomarker positivity is required for diagnosis, with evidence of AD related pathological changes not meeting the clinical criteria for MCI or dementia. Pre-clinical AD was further categorized into 3 stages mainly based on biomarker positivity. The reason for attempting to identify this stage is that it may provide a window of opportunity for drug trials which might be effective at this early disease stage (Sperling et al., 2011). MCI was defined by NIA-AA on the basis of both core clinical criteria (for clinicians) and research criteria (for clinical trials, including biomarker evidence). The MCI stage differs from the dementia stage mainly in its preservation of independence in activities of daily living (Albert et al., 2011). The core clinical criteria were sufficient to diagnose dementia due to AD in the NIA-AA guidelines. According to these criteria, dementia was defined as a progressive cognitive decline, diagnosed through history taking and which could not be explained due to other reasons. Moreover, the cognitive decline should be
sufficient to interfere significantly with activities of daily living, such as performing complex tasks at work, at home, acquiring and retaining new information, and deteriorations in language functions etc. Probable AD dementia as defined by NIA-AA was the same as that defined by NINCS-ADRDA but could have an amnestic or a non-amnestic presentation and with increased certainty among those carrying genetic mutations (amyloid precursor proteins (APP), presenilin 1 and 2). Possible AD dementia was defined among those having atypical AD or mixed dementia (McKhann et al., 2011).
The IWG initial guidelines identified various stages of AD including prodromal AD, typical AD, atypical AD, AD dementia, mixed AD and the pre-clinical state of AD. These guidelines emphasized the presence of biomarkers to diagnose different states of AD (Dubois et al., 2010). However, in their revised criteria, IWG emphasized the core clinical criteria in the diagnosis AD, so that their criteria could be used effectively in both clinical and research settings. In the revised guideline’s core clinical criteria, typical AD was defined as an amnestic syndrome of the hippocampal type. This syndrome can be identified in clinical settings by a decline in tests that assess effective registration of an item to be remembered and probe response to cueing as a measure of the storage abilities and associative function of the hippocampus. Biomarkers are a part of revised guidelines but now their purpose is to support the diagnosis rather being compulsory as in the previous IWG guidelines. The revised criteria also identified mixed AD, pre-clinical AD and atypical AD (Dubois et al., 2014).
2.3.2 Biomarkers of AD
Biomarkers are biological indicators that help to diagnose a disease with certainty. AD biomarkers were not available in 1984, when the first AD diagnostic criteria were formulated. The recently developed AD biomarkers have helped to increase the specificity of AD diagnosis and are included in both the IWG and NIA-AA criteria. A correlation between AD clinical symptoms and biomarkers has been demonstrated (Jack et al., 2010, Mormino et al., 2009, Perrin et al., 2009).
AD biomarkers are divided into two main categories: (1) biomarkers for amyloid deposition [CSF Aβ42, Positron emission tomography amyloid imaging], (2) biomarkers of neuronal injury [CSF total tau, phosphorylated tau, fluorodeoxyglucose positron emission tomography imaging, single photon emission tomography perfusion imaging, functional magnetic resonance imaging, hippocampal volume or medial temporal lobe atrophy by volumetric measures or visual rating].
Diagnostic guidelines of IWG and NIA-AA differ with respect to the use of biomarkers to diagnose pre-clinical AD (Sperling et al., 2011, Albert et al., 2011). In NIA-AA guidelines, a biomarker abnormality supported the AD diagnosis, but was not essentially required or sufficient; while IWG listed certain biomarkers as being required for AD diagnosis (Dubois et al., 2010, McKhann et al., 2011).
NIA-AA devotes equal importance to markers of Aβ deposition and neuronal injury in all stages of AD, while in the IWG-2 criteria, the presence of both markers i.e. decreased Aβ1-42 together with high total or phosphorylated tau in CSF, is essential for the diagnosis of typical AD (Dubois et al., 2014). Recently, the state and stage of AD have emerged as two different
concepts where state denotes a pathophysiological process (such as asymptomatic but at risk of AD) and stage refers to the progression of disease in a certain state (Dubois et al., 2016). AD has both a preclinical and a clinical stage; furthermore the clinical stage of AD comprises the prodromal and the dementia stages. The clinical phenotype of AD can be either typical or atypical. (Dubois et al., 2016).
Despite the major developments in biomarker identification, there is no consensus on whether they should be required for AD diagnosis (Herrup 2010, Pimplikar et al., 2010).
Moreover, the biomarker evaluation of AD pathology needs to undergo validation and standardization in terms of CSF collection, processing, performing quantitative assays, and accessibility and costs (Morris et al., 2014).
According to Diagnostic and Statistical Manual of Diseases 5th edition, AD is now classified as a major neurocognitive disorder and MCI as minor neurocognitive disorder based on the fact that it involves both neurological functions and interference with activities of daily living. It is true that diagnosing and labeling an individual with dementia without a definite diagnosis can raise serious ethical issues, on the other hand, it can help undiagnosed dementia cases who suffer from this disorder without proper care and support. Thus, their needs may be acknowledged and appropriate care can be planned after their diagnosis.
2.3.3 Midlife risk and protective factors of AD
AD is nowadays considered as a continuum of disease which begins well before the appearance of any clinical symptoms (McKhann et al., 2011). As a result, various protective and risk factors of AD have been recognized which operate throughout the individual’s life- span especially during midlife, thus affecting his/her likelihood of developing AD. Some of the protective and risk factors of AD pertaining to lifestyle and sociodemographic are listed in Figure 2 (Solomon et al., 2014a).
Risk factors of AD can be categorized into two main categories; modifiable and unmodifiable factors. Age, genetic constitution (APOE status) and family history of AD fall into the unmodifiable group of risk factors.
Age is the strongest predictor of AD, i.e. the risk of AD increases exponentially as the individual gets older. The presence of the APOE Ɛ4 allele is another well-established genetic risk factor for sporadic AD and >60% of AD cases carry at least one APOE Ɛ4 allele (Riedel et al., 2016). Women carrying one APOE Ɛ4 allele had similar risk of AD as men homozygous for APOE Ɛ4 (Farrer et al., 1997). The APOE gene encodes a protein which acts as a major component for central nervous system lipoproteins and is thus involved in lipid transport in brain (Manaye et al., 2013). The APOE ε4 isoform increases the risk of AD through increased production of amyloid Aβ, and a decrease in dendritic spine density (Rodriguez et al. 2013, Dumanis et al., 2009).
Another important determinant with respect to APOE ε4 status is its interaction with female sex steroid hormones. APOE is a biological factor which associates with sex, genetic, and lifestyle related factors (education, physical activity, smoking, occupation status, and job situation) to alter AD-related pathology (Rocca et al., 2014b). Women homozygous for APOE Ɛ4 were found to have lower CSF Aβ levels in a dose response manner in late onset AD but not in early onset AD (Mehrabian et al. 2015). In females, the influence of APOE Ɛ4 presence was more pronounced on the neuropsychiatric symptoms of AD (Xing et al., 2015), and APOE Ɛ4 carriers showed more severe amyloid pathology on positron emission tomography than was present in non-carriers (Jack et al., 2015).
The modifiable risk factors of AD include cerebrovascular and cardiovascular risk factors, such as obesity, smoking, alcohol intake, and high fat diet (Solomon et al., 2014a) with higher education, social and physical activity, and doing a mentally stimulating job being associated with a decreased risk of dementia (Wilson et al., 2007). The mechanisms through which higher education, higher socioeconomic status, being socially active and doing a mentally stimulating job protect from AD are possibly mediated through the increase in cognitive and brain reserve (Wilson et al., 2010, Fotenos et al., 2008). A higher brain reserve would enable a brain to better withstand pathological insults due to the presence of the larger number of healthy neurons, while cognitive reserve denotes the brain’s ability to exploit alternative networks of brain to combat the developing pathology, such as in AD (Sperling et al., 2011). Both cognitive and brain reserve enable the brain to tolerate the initial symptoms of dementia without showing any clinical symptoms for a longer duration of time, but it may also lead to a more rapid decline once compensatory mechanisms stop functioning (Fotenos et al., 2008). APOE Ɛ4 and education influence the onset of dementia
independently as well as interactively i.e. the risk of dementia is halved among APOE Ɛ4 carriers with high education in comparison to APOE Ɛ4 carriers with low education (Wang et al., 2012).
Smoking, alcohol intake, low physical activity, and high fat diet etc. predispose to higher risk of AD. These habits may promote accelerated aging of brain through metabolic and glucose dysregulation, oxidative stress and chronic inflammation, all of which are factors that increase the risk of AD either independently or operate through increasing the risk of cardiovascular and cerebrovascular diseases (Arvanitakis et al., 2004, Yaffe et al., 2009, Schmidt et al., 2002). Volunteering in any task has been associated with decreased mortality among older people, and the underlying reason can be that volunteerism stems from emotional wellbeing and as well as the social interactions gained by this activity (Harris and Thoresen et al., 2005). Obesity itself is an important predictor of AD. Midlife BMI of >30 in conjunction with high systolic blood pressure and higher total cholesterol increase the risk of AD either through the inflammation associated with obesity or increasing the risk of the metabolic syndrome (Kivipelto et al., 2005). Diabetes mellitus is another risk factor acting either independently or in combination with obesity to increase the risk of AD via insulin resistance and microvascular disease in brain. Brain insulin production may be inhibited by peripheral hyperinsulinemia, which in turn may decrease the clearance of amyloid from the brain (Barnes and Yaffe 2011). Moreover, adipocytes are known to secrete various hormones (leptin, cortisol) and cytokines (tumor necrosis factor-alpha and interleukin 6) which collectively increase the risk of AD by inducing inflammation in brain and by altering brain beta amyloid levels (Profenno et al., 2009). An elevated serum cholesterol level is a well- established risk factor for AD since it promotes the formation of amyloid beta in neuronal cell membranes through the formation of cholesterol rich areas which preferentially process APP into Aβ (Casserly and Topol 2004). High blood pressure in midlife is associated with a higher risk of late-life dementia. Hypertension increases the risk of developing white matter lesions, small and large vessel disease, and brain atrophy, all of which may converge and thus link the higher risk of dementia with high blood pressure (Launer et al., 2000).
Hypertension also affects the endothelial lining of blood vessels, altering their permeability and inducing proinflammatory and procoagulant responses in cell membranes, which in turn may trigger the formation of neuritic plaques, a hallmark of AD (Hallenbeck 1994). A decline in the incidence of dementia with antihypertensive drugs among older people support the view that hypertension is a modifiable risk factor for dementia (Forette et al., 2002, Feigin et al., 2005).
Other important factors associated with a decreased risk of AD are higher intellectual activity, being in a relationship (marriage), which also delays AD through increasing cognitive reserve (Vemuri et al. 2014, Sundstrom et al., 2016). In summary, preventing or delaying the onset of AD clearly demands monitoring and modifying of midlife risk and protective factors for AD (Solomon et al., 2013).
With regard to the several protective factors, hormone therapy (HT) holds a special place as a potential therapeutic to prevent or delay onset of dementia in women. Considering the longer life span of women, the menopause also marks a midlife event (mean age 51 years) as more than one third of a woman’s life span is spent in the postmenopausal state. Thus,
the higher risk of AD among females than males can be associated biologically with the decline in the amounts of sex steroid hormones (estrogen and progesterone) at menopause (Vest and Pike 2013). The use of conjugated equine estrogens was not associated with a cognitive decline in a recent meta-analysis, but this report did not consider other formulations of estrogens such as estradiol. In the same meta-analysis, cognitive training was associated with a decreased risk of cognitive decline, while current smoking and APOE Ɛ4 genotype were associated with an increased risk of AD (Plassman et al., 2010). The biological effectiveness of estrogen is reduced in the presence of the APOE Ɛ4 allele (Manaye et al., 2013). It has been reported that estrogen conferred protection against cognitive decline among APOE Ɛ4 negative women but not in APOE Ɛ4 positive women (Yaffe et al., 2000a).
This concept may also explain the higher risk of AD among women which might be mediated through the APOE interaction with female sex per se.
Though no clear guidelines are available whether or not to use HT among postmenopausal women as a means to prevent dementia or AD, much research has been conducted in this field during the past two decades; there is evidence of a neuroprotective potential of HT emerging from experiments conducted in animals as well as in observational trials. Some discrepancies have also been seen with respect to the clinical trial findings; these will be discussed in detail in the following chapters.
Figure 2: Risk and protective factors of AD (Solomon et al., 2014) 2.3.4 Sex based dimorphism in brain and AD
Of the 5.2 million people with AD in United States, 3.3 million (two thirds) are women (Alzheimer's Association 2016). Similarly, in global terms, more than 60% of patients with AD are women (Riedel et al., 2016). The higher risk of AD among women than men can be simply attributed to their longer life span (Gao et al., 1998) but it may also be linked to underlying sex-based dimorphism in the human brain (Woods and Tsui 2014). Sex and
RISK FACTORS
Unmodifiable Age
Family history APOE Ɛ4
Modifiable
Cerebrovascular risk factors Cardiovascular risk factors Smoking
Alcohol intake High fat diet
PROTECTIVE FACTORS
Education
Social activities
Cognitive engagement cognitive reserve Physical activity
Mentally stimulating job
High socioeconomic status
Mediterranean diet Antioxidant vitamins NSAIDS
Statins
Hormone therapy Polyunsaturated fatty acids
gender have emerged as separate entities recently, where sex defines biological characteristics such as chromosomal constitution (XX or XY), gonadal and hormonal differences, while gender refers to cultural, psychological and social differences (access to education, occupation) between men and women (Ristvedt 2014, Mielke et al., 2014). Thus, identification of both sex- and gender- based risk and protective factors are critical in understanding a chronic illness such as AD. Sexual dimorphism is driven by the levels of the sex-specific hormones prevailing during prenatal period, adolescence, puberty, and adulthood (Li et al., 2014) and these are governed by hormone dependent gene activations in a sex specific manner (Nugent et al., 2012). One such example is the sex hormone mediated language development which differs depending on whether the postnatal hormonal surge is mediated by estrogen or testosterone (Schaadt et al., 2015).
The distinctive distributions of estrogen and androgen receptors in brain account for differences in performance in brain tasks (Li and Singh 2014); this has been linked with certain pathological variants; long term estrogen depletion was reported to be associated with cognitive decline (Mielke et al., 2014); excess hormone exposure results in polycystic ovarian syndrome in females (Nugent et al., 2012); and X-inactivation is associated with an increased AD risk among females (Ferrari et al., 2013).
All of these mechanisms suggest that sex-specific hormones affect an individual’s likelihood of developing AD in a variety of ways. It can be mediated through down-regulation of estrogen receptors in hippocampus in the case of long term-estrogen depletion, thus affecting the main area of the brain involved in neuroprotection and cognitive enhancement (Mielke et al., 2014). There are also possible indirect pathways e.g. the higher risk of AD among women with polycystic ovarian disease, who also have a higher risk of developing the metabolic syndrome due to insulin resistance, higher BMI and cholesterol levels. These metabolic changes predispose women with polycystic ovaries to a higher risk of AD (Nugent et al. 2012). Similarly, genetic mechanisms may be involved in the higher prevalence of AD among women, such as inactivation of the X-chromosome during embryogenesis since this chromosome mainly carries neuroprotective genes, or it can be due to unknown non-genetic and epigenetic mechanisms (Ferrari et al. 2013).
Moreover, sexual dimorphism in brain is associated with a higher incidence of AD among women through sex-specific white matter lesions (Gallart-Palau et al., 2016) and an increased rate of cognitive decline among females than males (Laws et al., 2016, Koran et al., 2016). These sex-specific associations involve a complex interplay between hormonal, genetic, and environmental factors (Carter et al., 2012).
2.4 FEMALE SEX STEROID HORMONES
Primary female sex-specific hormones are estrogen and progesterone. They are small, hydrophobic molecules carried through serum globulin in the bloodstream. Sex hormone production is regulated by the hypothalamic pituitary gonadal axis through tightly controlled hormonal and neural signals between the central nervous system, the pituitary and the ovaries respectively. The following hormones are part of the hypothalamic pituitary gonadal axis: 1. Gonadotrophin-releasing hormone; 2. Gonadotrophins i.e. luteinizing
hormone and follicle stimulating hormone (Henry, Norman 2014). Low levels of circulating estrogens and progesterone trigger the release of gonadotrophin releasing hormone from hypothalamus which then stimulates the pituitary to release follicle stimulating hormone and luteinizing hormone, both of which act on the developing follicles to produce estrogens and progesterone (Henry and Norman 2014, Blair et al., 2015b).
2.4.1 Estrogens and progesterone
Estrogen is produced mainly in the reproductive organs (ovaries) during reproductive life and from non-reproductive sites (liver, brain, bone, adipose tissue, muscle, heart) before puberty and after menopause though its level are rather low in comparison with the ovarian estrogens. There are 3 types of estrogen endogenously present in females: estradiol (17β estradiol), estrone, and estriol. Estradiol is the most potent and most prevalent during the reproductive years; estrone is synthesized from adipose tissue mainly after menopause, while estriol is prevalent during pregnancy and is produced by the placenta (Cui et al., 2013).
Sex hormones exert their biological actions through cell and tissue specific receptors. The concept of the estrogen receptor was proposed in the late 1950s (Jensen and Jacobson 1962) and the estrogen receptor α was identified in the 1960s (Toft, Gorski 1966). The gene for human estrogen receptor α was cloned in 1986 (Greene et al., 1986) and that for estrogen receptor β was discovered in the late 1990s. Both receptors are widely distributed in brain and body (Henry and Norman 2014).
Estrogen exerts its specific effects primarily through its nuclear receptors i.e. estrogen receptor α or β (genomic), but also by binding to its membrane bound receptors (non- genomic) (Hewitt et al., 2016, Blair et al., 2015b). The estrogen receptor α is primarily expressed in reproductive organs whereas estrogen receptor β is expressed in a wide variety of tissues. Both estrogen receptors are present in various brain regions including amygdala, cortex, hippocampus, hypothalamus, substantia nigra, stria terminalis, and preoptic area (Cui et al., 2013).
The other important female sex steroid hormone is progesterone, which in reproductive life is produced in the ovaries (corpus luteum), placenta, and adrenal gland (cortex). There are two types of receptors for progesterone, i.e. progesterone receptors A and B; the B-form of the progesterone receptor is more potent than the A-form (Henry and Norman 2014). The activity of the progesterone receptor is dependent on estrogen as well as the properties of the estrogen receptor in target tissues; progesterone receptors also act as a transcription factor similarly as the estrogen receptors. Estrogen receptors are more widely distributed than progesterone receptors, which are limited to uterus, ovary in the periphery, but they are also present in several areas of the brain including pituitary. The presence of progesterone receptors throughout brain means that it should not simply be considered as a reproductive hormone.
2.4.2 Mechanisms of neuroprotective effects of estrogen and progesterone
In vivo animal and human autopsy studies have demonstrated the wide distribution of the estrogen receptor α throughout the hypothalamus, forebrain, and amygdala (Osterlund and
Hurd 2001, Mitra et al., 2003) and the estrogen receptor β in hippocampus and cerebral cortex (Mitterling et al., 2010, Ostlund et al., 2003) emphasizing the role of estrogen in brain functions. A brief summary of some of the neuroprotective mechanisms of estrogen and progesterone is presented in Table 1.
Estrogen can confer neuroprotection through induction and regulation of synaptic activity in hippocampus and the formation of excitatory synapses through N-methyl-D-aspartate receptors (McEwen and Alves 1999, Hao et al., 2003, Jacome et al., 2016) and estrogen receptor mediated gene transcription to regulate hippocampal function (Han et al. 2013).
Estrogen may increase synaptic plasticity whereas progesterone antagonizes this protective effect of estrogen (Baudry et al. 2013) and increases neuronal spine density in prefrontal cortex and hippocampus (Tang-Wai et al. 2004, Shanmugan, Epperson 2014).
Estrogen enhances hippocampal learning through neurogenesis although this may depend upon the type, duration, and time of initiation of HT. Estrogen increases neurite growth and synaptic plasticity among healthy neurons, but not in aged neurons which are less responsive to estrogen. Moreover, a longer duration of naturally occurring estrogen (estradiol) use encompassing the critical time period when neurons are still healthy is another important factor determining the effect of HT on cognition (Duarte-Guterman et al., 2015, Brinton 2008c, Rapp et al., 2003a).
There are several major pathways through which estrogen may exert neuroprotection; up- regulation of cholinergic activity in basal forebrain and hippocampus, especially when neurons are healthy (Gibbs and Aggarwal 1998, Gibbs 2010, Newhouse and Dumas 2015);
increasing the level of neprilysin (an enzyme which degrades amyloid β) (Huang et al., 2004); acting as an antioxidant in brain; and increasing blood flow and glucose transport in brain (Brinton 2008b, Rettberg et al., 2014); decreasing tauopathy (Grimm et al., 2016);
decreasing levels of pro-inflammatory cytokines and also by attenuating the stress induced levels of glucocorticoids (Shivers et al., 2015, Herrera and Mather 2015); it can also activate prefrontal cortex and medial temporal lobe, thus improving cognition (spatial working memory, object recognition, and reference memory) (Rapp et al., 2003a, Markowska and Savonenko 2002).
Estrogen mediated neuroprotection in cognitive tasks involving white matter (such as reasoning, learning, planning) is selective (Pompili et al., 2012) i.e. estrogen related neuroprotection is mediated by an increase in levels of brain derived neurotropic factor and nerve growth factor as well as improving glucose metabolism and cerebral blood flow and diminishing the amounts of free radicals through its anti-oxidant properties (Monk and Brodaty 2000).
