Kenneth Kwok Chi Yau

A thesis submitted in conformity with the requirements for the degree of Master of Science

Graduate Department of Pharmacology University of Toronto

O Copyright by Kenneth Kwok Chi Yau 2000

National Library I * m of Canada Bibliothèque nationale du Canada

Acquisitions and Acquisitions et Bibliogaphic Services services bibliographiques

395 Wellington Street 395. rue Wellington OttawâON K1AON4 OttawaON KIAON4 Canada Canada

your nle votre referame

Our n19 Notre feférence

The author has granted a non- exclusive licence dowing the National Library of Canada to reproduce, loan, distribute or sell copies of this thesis in microform, paper or electronic formats.

L'auteur a accordé une licence non exclusive permettant à la Bibliothèque nationale du Canada de reproduire, prêter, distribuer ou vendre des copies de cette thèse sous la forme de microfiche/film, de reproduction sur papier ou sur format électronique.

The author retains ownership of the L'auteur conserve la propriété du copyright in this thesis. Neither the droit d'auteur qui protège cette thèse. thesis nor substantid extracts from it Ni la thèse ni des extraits substantiels may be printed or otheMrise de celle-ci ne doivent être imprimés reproduced without the author's ou autrement reproduits sans son permission. autorisation.

Assessing and Predicting Treatment Responses to Cholinesterase Inhibitor

Pharmacotherapy in Alzheimer's disease. Kenneth Kwok Chi Yau, MSc 2000, Department of

Pharmacology, University of Toronto

Cholinesterase inhibitors (ChEIs) are the only pharmacotherapy marketed for Alzheimer's

disease (AD). This work evaluated the potential utility of a penpheral cholinergic probe, which

previously detected age-related differences in central cholinergic function, in AD (Study 1 ) and

assessed the efficacy and safety of oral ChEIs in AD (Study 2).

Elderly controls (n=4) showed a 33.5+6.0% decrease in pupil size from a pilocarpine challenge

test (PIL-CT) compared to 25.6*7.3% for AD patients (n=6). In AD patients (n=5) a 4.821 -6-

point increase in ADAS-cog score aRer 12 weeks of donepezil treatment was accompanied by a

13.3k7.3% decrease in PIL-CT. End of treatment outcornes and pre-treatment PIL-CT responses

didn't correlate. Changes in ADAS-cog scores and PIL-CT did correlate in certain subjects.

A meta-analysis of oral ChEIs in AD revealed a therapeutic effect of 12%. ChEIs had a yreater

incidence of adverse events (7%). dropouts ( IO%), and dropouts due to adverse events ( 1 1 )

over placebo.

1 would like to acknowledge the foilowing people who have supported me and contributed to the

to completion of this thesis:

Dr. Claudio A. Naranjo. my supervisor, for taking me on as your student and for always

çhalienping me to open my mind and consider possibilities that 1 othenvise would not have.

Dr. Nathan Hemnann, my advisor, for always making time to listen to me and for always

believing in me. You have been a tower of strength and wisdom and a wellspring of

encouragement and hope. Thank you for going to bat for me when 1 needed your support.

Dr. Krista Lanctôt. my CO-advisor, for taking me under your wing and for being like an older

sister to me. Thank you for al1 your guidance, support, and spreadsheets. but most of al1 for

always being there to listen to me when I needed a fhend.

Dr. Usoa Busto (Intemal thesis reviewer), Dr. Kenneth Shulman (Extemal thesis reviewer).

and Dr. Richard Rozmahel (Oral Examination Chair), for your time and effort spent

reviewing my thesis and attending my defense.

Diana Clark. for everything you have done to make sure this thesis got done on time. 1

would have been lost, or at least very conhised. without a11 your help.

Dr. Catherine Birt and Mary Ghazalian, for screening rny subjects, and Ingrid Quinton, for

preparing the eye drops for my study, no matter how little notice 1 provided.

Dr. Beth Sproule, for always making time for me whenever 1 popped by with a question and

for preparing the meds for my study, no matter how linle notice I provided.

Dr. Barbara Liu, for donating your time for speedy and accurate ratings on the meta-analysis.

0 Dr. John Iazetta and William Cornish, for your pharmacy support.

Ruth Croxford, for making time to provide me with statistical guidance and advice, no matter

how little notice 1 provided.

Betty Rychlewsky, Sharon Crosgrey, Irene Estock, and Dilshad Ratansi for your fnendships

and for helping me to track down Dr. Hemnann and Dr. Shulman.

Dr. Bill Gentiles, John Lee, and Teny Tang, for their assistance in repairing my pupillometrr

and finding me a new one when the other when went down.

Linda Tatoff. For taking care of al1 the little things for me and for al1 your suppon and

encouragement.

Della (Obi-wan) Knoke, for sharing with me not only your wisdom about research and

everything eise, but for letting me look out your window whenever 1 dropped by.

Alan Chu. Edward Sturman III, Jemifer Glass, Judi! Hall (my proof-reader), Kemal Kilic.

Laura Cardenas, Lescia Tremblay, Lyla Khan, Paolo Mazzotta. Shehnaz Moosa. and Susy

Felix, for a11 laughs and srniles, for al1 support and encouragement. and for al1 the mernories.

Thank you for making the Psycho F m a wondefil, heart-warming experience.

Bettina Hsueh, my Sunshine, for being with me through al1 the ups and downs these past few

years. Thank you for taking this joumey with me each step of the way.

Hubert and Irene Yau, my parents, Jemifer Yau, my sister, and Paul and Wimie Yau. my

grandparents for your unconditional support throughout al1 my years of education.

So grandma, this is what I've been doing for the past 2 years.

Finally I would like to thank God for granting me the peace, patience, and perseverance to

see this through to the end. '4 can do al1 things through Him who gives me strength"

(Philippians 4: 13)

Hey everyone, look what 1 did! See you out there. God Bless.

iv

... ACKNOWLEDGEMENTS ............................................................................... 111

............. INTRODUCTION ....... ............................................................................................. 1

STUDY 1 . A PENPWERAL CHOLINERGIC PROBE TO MONITOR ALZHEIMER'S DISEASE AND PREDICT TREATMENT RESPONSE TO CHOLINESTER4SE

...................................................................................... INHIBITOR PHARMACOTHERAPY 3

Introd~tction: ~ h e role of ceniraf cholinergic -rem in memon c ~ n d cognition ..................... 3

........................................................... The tietrrocheniistry of the cenfral cholittergic system 3

...................................... Choli~iergic receptor subtypes: nittscarinic and nicotinic receptors i

...................................................................... Funcrion and location ofchohtergic receptors .i

................................................................ Neuropathology of AD: the choiinergic hypoihesis 6 .

nie Cholinergie Hyporhesis: Not the whole stov in AD ........................................................

A peripheval clioiinergic probe as a mode1 for meastiring central cholinergie ncrivif~ ......... Y

....................................... The physiology of the penpheral cholinergie system of the iris 8

...... Previous use of cholinergie and anticholinergic eye drops: a diagnostic tool for AD? S

................................................ A Tropicarnide-Pilocarpine challenge test (TROPIL-CT) 1 1

A Pilocarpine challenge test (PIL-CT) ............................................................................. 11 Current Directions ............................................................................................................. 13

.......................................................................... Cholinergie Replacement Therapv for rLD 13

...................................................................................................... Cholirtesterase in hibitors i 4

............................................................................. ............................... HYPOTHESIS 1 -1 ....... 1 6

.......................................................................................................... METHODS FOR STUDY 1.1 1 6

Studv Design ......................................................................................................................... 1 7

................................................................................................................. Study medications IS

........................................................................................... Meastirement of Pupil Diameter 18

........................................................................................................ Instillation of eye drops i 9

Safey ................................~~................................................................................................. 21

Data Ana(vsis ........................................................................................................................ 21

RESULTS FOR SWDY 1.1 .......................................................................................................... 23

Sztbjecfs ................................................................................................................................. 23

Pilocarpine effect on pzipif size ....................... .. ................................................................ 23

HYPOTHESIS 1.2 ......................................................................................................................... 26

METHODS FOR STUDY 1.2 .......................................................................................................... 26

Su bjecrs ................................................................................................................................. 26

Stzidv Design ......................................................................................................................... 26

Trea [men t Response Measzirernents ...................................................................................... 28

Safety ................................................................................................................................... 29

Data .4 na[\sis ....................................................................................................................... 30

RESuLTS FOR STUDY 1.2 ......................................................................................................... 31

Sit bjects ................................................................................................................................. 31

Eficnqv Assessrnen t .............................................................................................................. 32

.............................................................................................. Primary Outcome Measures 32

.......................................................................................... Secondary Outcome Measures 3 2

PIL-CT as a predictor of Ch El response .............................................................................. 35

................................................................................... PIL-CT as a meusure of disease state 38

............................................................................................................................... D[scussio~ 40

............................................................................................ CONCLL~SIONS 56

STUDY 2 . EFFICACY AND SAFETY OF ORAL CHOLINESTERASE INHIBITORS IN .a ZHEIMER'S DISEASE ...........................................................................m............................ 57

............................................................................................................................ BACKGROUND 57 . -. ........................................................................................................................ Meta-ana&sts r i

................................................................................................... METHODS 5s

.................................................................................................................. Literaiztre Search j 8

.............................................................................................. Incltcsion/Excltîsion Criteria 58

............................................................................................................ Assessrnent of Studies 59

................................................................................................... Assessrnent of Inclusion 59

Quality Assessrnent ........................................................................................................... 60

Data Extraction ............... ................................................... ............................................. 60

Data Analysis .................................................................................................................... 61

RESULTS .................................................................................................................................... 65

Stttdies .......................................................................................... ..................................... 6 5

Comparative analyses ........................................................................................................ 65

E ficacy Analyses ............................................................................................................. 65

................................................................................................................. Safety Analysis 73

7'6 ............................................................................................................... Si jtgk il rni Anal-vsis

Analysis by Dose .............................................................................................................. 76

Analysis by Class ............................................................................................................. S3

Quality Assessrnent ............................................................................................................... $3

DISCUSSION ............................................................................................................................. S4

CONCLUSIONS ............................................................................................. 89

RECOMMEND ATIONS ............................................................................................................ 90

TABLE f . CHARACTERISTICS OF ELDERLY AMI AD SUBECTS ........................................ 24 * 1 .......................................................... . TABLE 2 SUMMARY OF OUTCOME MEASURES JJ

............................................. TABLE 3 . SUMMARY OF CHEI DOSAGE CLASSIFICATIONS 64

TABLE 4- STUDIES MEETING INCLUSION CRiTERL4 FOR META-ANALYSE OF O R i L CHOLINESTERASE

INHIBITORS Pi ALZHEIMER'S DKEASE ............................................................ 66

TABLE 5 - DATA E X T M C E D FROM STUDIES INCLUDED IN META-ANALYSE OF ORAL

...................................... CHOLINESTERASE INHIBITORS IN ALZHEIMER'S DISEASE 70

vii

TABLE 6 . SUMMARY OF COMPARATIVE OUTCOMES FOR .4 LL D.4TA EXTR4CTED INCLCDING

QUALITY ANALYSES FROM META-ANALYSIS O F CHEI TREATMENT IN AD .................. 74

. TARLE 7 SUMMARY O F COGNITIVE EFFICACY DATA BY DOSE AND CLASS .......................... 77

. TABLE 8 SUMMARY O F GLOBAL ASSESSMENT DATA BY DOSE AND CLASS ......................... 78

. .......................................... TABLE 9 SUMMARY O F SAFETY DATA BY DOSE AND CLASS 79

TABLE 10 . SUMMARY OF DROPOUT DATA BY DOSE AND CLASS ...................................... SO

. ......... TABLE 1 1 SUMMARY O F DROPOUT DUE TO ADVERSE EVENT DATA BY DOSE .A ND CL.L\SS 81

TABLE 12 . SUMMARY O F DROPOUT DUE TO ADVERSE EVENT DATA BY DOSE FROM STCDIES

INCLUDED IN EFFICACY ANALYSES ............................................................... 82

FIGURE 1 . NEUROCHEMICAL ORGANIZATION OF THE CHOLINERGIC SYNAPSE ........................ 4

FLGURE 2 . [NNERVATfON OF THE PUPILLARY MUSCLES ................................................... '1

FWRE 3A . SCHEMATIC D[AGRAM O F UNILATEIUL PUPILLOMETER ................................. 20

FIGURE 3B . SCHEMATIC D[AGRAM OF BILATERAL PUPILLOMETER ................................... 20

FIGURE 4- TIME COURSE OF PUPILLARY RESPONSE T O PILOCARPtNE IN ELDERLY

................................................................................. .i\ ND AD PATIENTS 25

. .............................................. FIGURE 5 - S U ~ C ~ ~ M A R Y OF STUDY DESIGN FOR STUDY 1 2 27

FIGURE 6 - SUMMARY OF TRlAL PROFILE FOR STUDY 1.2 .............................................. 31

FIGURE 7A - COMPARISON OF MEAN BASELINE A N D POST-DONEP EZIL PIL-CT RESPONSE ......... 34

FIGURE 7B - COMPARISON OF MEAN BASELINE AND POST-DONEPZIL ADAS-COG SCORE ......... 34

FIGURE 8 - SUMMARY O F CORRELATIONS BETWEEN BASELINE PIL-CT RESPONSES .A ND

............................................ TREATMENT RESPONSE TO DONEPEZIL TREATMENT 36

FIGURE 9 - SUMMARY OF PL-CT RESPONSES FOR RESPONDERS AND NON-RESPONDERS T 0

........................................................................... DONEPEZIL TREATMENT 37

F~GURE I O - SUMMARY OF CORRELATiONS BETWEEN PIL-CT RESPONSE

........................................................................... AND ADAS-COG SCORE 39

.............................................. FIGW f 1 - COGNITIVE THERAPEUTIC EFFCTT OF CHEIS 75

FIGURE 12 - GLOBAL ASSESMENT THEEUPEUTIC EFFECT OF CKEIs ................................. 75

APPENDIX A . CONSENT FORM FOR STUDY 1 . I ...................................................... 1 1 1

APPENDIX B . CONSENT FORM FOR STUDY 1.2 ........................................................ 1 12

!~PPENDIX C . INCLUSION FORM FOR META-XNALYSIS ................................................ 1 13

APPENDIX D . QUALKY ASSESSMENT FROM FOR META-ANALYSIS ................................. 1 14

. APPENDIX E DATA EXTRACTION FORM FOR META-ANALYSIS ...................................... 1 15

A P P ~ D L X F . META-A~AL'ISLS SPREADSHEE'TS FOR COMPARATIVE CALCULATIONS ............. 1 16

APPENDIX G . META-ANALYSIS SPREADSHEETS FOR SINGLE-ARM CALCULATIONS

BY CHEI DOSE .................................................................................... 1 17

APPEND~X H . META-ANALYSIS SPREADSHEETS FOR SINGLE-ARM ChLCLiL.ATIONS

BY CHEI CLASS ................................................................................... 1 1s

Alzheimer's disease (AD) is the most common neurodegenerative disorder in the western world

(1). The prevalence is considered to be less than 5% for people 65 years of age or less but is

estirnated to double every 5 yean between the ages of 65 and 85 (2) to as high as 5096 of people

over 85 with probable AD (3). Neuropathologically. AD is characterized by widespread central

cholinergic deficits in areas of the brain known to be important in memory and learning such as

the cortex and hippocarnpus (4-6). These deficiencies include neuronal darnage and decreases in

the cholinergic neurotransmitter acetylcholine (ACh) (7-9). Central cholinergic impairment has

been well documented in conditions associated with disruptions in cognition such as Dow's

syndrome (10). The cholinergic deficits observed in AD were proposed as the cause for the

deterioration of cognition seen in this disease ( I 1).

Cho linergic replacement has been studied as a treatment for AD; however. only cholinesterase

inhibitors (ChEIs) have been marketed for pharmacotherapy (12. 13). These compounds

increase cholinergic neurotransmission by blocking acetylcholinesterase (AChE). the enzyme

which breakdowns down ACh (14-16). Tacnne was the first ChEI approved for the treatment of

AD, but was plagued by a high incidence of adverse events, in particular elevated liver enzymes

that required a large proportion of patients to discontinue treatment (17). Newer ChEIs. like

donepezil, are reportedly similar in efficacy to tacnne, but considerably better tolerated ( 12. 13.

18). Clinical trials of ChEIs consistently report that these medications are modestiy efficacious

for the treatment groups as a whole, however, they appear to be ineffective for many individuals

(19-22). This variability in response suggests that further clarification of the efficacy and safety

of ChEIs is necessary. In addition, predictors of treatment response may have an important role

in selecting suitable candidates for trials with ChEIs.

Currently, there is neither a simple, non-invasive diagnostic test for AD, nor any objective means

to monitor disease progression or predict treatment response to ChEIs. in 1994. Scinto et al.

reported that AD patients demonstrated a hypersensitive pupillary response to the cholinergic

antagonist tropicamide and proposed that it may be used as a diagnostic test For AD ( 2 3 ) . .A

number of groups failed to replicate those results (21). Our group showed that the choliner$c

agonist pilocarpine detected age-related differences in central cholinergic function between

healthy young and elderly subjects (25).

This thesis is proposed as a logical extension of our previous work. Firstly. the utility of a

penpheral probe to measure central cholinergic hnction and to predict ChEI treatment response

was assessed (Study 1). Secondly, a quantitative evaluation of the efficacy and safety of ChEI

phmacotherapy in the literature was performed (Study 2). Study 1 was divided into 2

substudies that were performed concurrently. The studies included in this thesis are as follows:

Study 1: A Peripheral Cholinergic Probe to Monitor Alzheimer's disease and Predict

Treatment Response to Cholinesterase Inhibitor Pharmacotherapy

Study 2: Effkacy and Safety of Oral Cholinesterase Inhibiton in Alzheimer's disease

Each of these studies represents a unique aspect of this work and are presented in sequencc.

Study 1 is presented with a unified background, discussion, and conclusions. and substudy-

specific hypotheses, methods, and results. Study 2 is presented individually with a study-specific

background, methods, results, discussion, and conclusion. The thesis concludes with

recommendations encompassing the entire project.

STUDY 1 - A PEIUPEIERAL CHOLINERGIC PROBE TO MONITOR ALZHEIMER'S DISEASE AND PREDICT TREATMENT RESPONSE TO CHOLINESTERASE INHIBITOR PHASMACOTHERAPY

Introduction: the role of central cholinergic system in memory and cognition

The central cholinergic system has been well studied due to its involvernent in memory and

cognition (26-3 1). Studies involving both human and animal models have shown that centrally

acting cholinergic agonists and antagonists augment and impair cognitive functioning.

respectively (26, 28,32-36). Changes in the cholinergic system occur as a result of normal aging

and disease. Decreases in central cholinergic Functioning are seen in healthy elderly individuals.

leading to cognitive decline including difficulties with storage and retrieval of memory (34. 5 7 .

3 8). Studies have found decreased cholinergic precunor uptake and neurotransmitter synthrsis

and release, as well as changes in receptor ligand binding, in the brains of aged rodents in

comparison to those of young animals (3941). Decreased central cholinergic function has also

been observed in disease states that exhibit deficits in cognitive and intellectuai functioning. such

as Alzheimer's disease, Parkinson's disease, Down's syndrome, and the alcoholic Konakoffs

syndrome ( 10,424).

The neurochemistry of the central cholinergic systern

Cholinegic neurons derive their narne fkom the primary neurotransmitter they release frorn their

nerve terminah, acetylcholine (ACh) (45). ACh is synthesized in the cytoplasm by the enzyme

choline acetyltransferase (ChAT) fkom acetyl-coenzymeA (acetyl-CoA) (produced in the

mitochondria) and choline (imported kom the extracellular fluid). Once formed, ACh is stored

in small membrane bound vesicles concentrated near the synaptic clefi. When the nerve terminal

is depolarked by a sufficient action potential, the in f lu of calcium ions tnggers a cascade of

events resulting in the Fusion of vesicles containing ACh with the terminal membranes. The

neurotransmitter is then released into the synaptic clefi where ACh binds to and activates pre-

synaptic and post-synaptic cholinergic receptors. The enzyme acetylcholinesterase (AChE)

terminates the action of ACh by breaking it down into its component parts. acetyl-CoA and

choline (Figure 1) (45-47).

ACh n

Figure 1 - Neurochemical organization of the cholinergic synapse.

Abbreviations: AcCoA = acetyl-CoA; ChAT = choline acetyltransferase; ACh =

acetylchohe; AChE = acetylcholinesterase; ca2' = calcium ions; ~ a ' = sodium ions.

Cholinergic receptor subtypes: muscarhic and nicotinic receptors

Cholinergic neurotransmission is rnediated by two main receptor subtypes: muscarinic and

nicotinic cholinergic receptors. These subtypes are both smicturally and pharrnacologically

different. The nomenclature stems fiom the alkaloids muscarine and nicotine. which. through

differential stimulation, originally identified the two subtypes (45). Muscarinic receptors are

membee of the G protein linked "serpentine" fmily of recepton, consisting of 7 hslically

arranged transmembrane domains coupled to G proteins through the third cytoplasmic loop.

There are 5 subtypes of muscarinic recepton that have been cloned and named according to their

order of identification (ml. m?, m3, mr, and m5). These subtypes share over 9Oo6 homology in

the transmembrane domains. but differ significantly in the intracellular and extracellular loops

and tails (48). Stimulation of the muscarinic receptor by an agonist (e.g. muscarine. XCh) results

in the activation of G proteins that either stimulate (ml, m,, r nc ) or inhibit (m:, m ~ ) signal

transduction events (e.g. enzyme activation) further downstream (49). Nicotinic receptors are

cy 1 indrical cation-selective ion channels consisting of 5 polypeptide subuni ts. .4gonist ( e . ~ .

nicotine, ACh) binding results in conformational changes in the pentameric molecule allowing

the transient passage of sodium ions into the ce11 fi-om the extrace1lula.r fluid, ultimately leading

to depolarization and potentiation of a signal (45,46).

Function and location of cholinergic receptors

Cholinergic receptors are ubiquitous and play an integral part in the autonornic nervous system

(45). Nicotinic recepton are typically found in autonomic ganglion cells and the neuromuscular

junctions of skeletal muscle. Also important centrally, nicotinic sites are primarily detected in

the spinal cord (45, 46). In contrast, peripherally, muscarinic receptors are located only at

effector sites including cardiac and smooth muscle, glands, and endothelium. In the central

cholinergic system. muscarinic receptors are found rnostly in the brain (45, 46). More

specifically. muscarinic cholinergic neurons arise in the basal forebrain (nucleus basalis of

Meynert, diagonal band oFBmca, and the media1 septum) and innervate the cortex. including the

hippocampus, arnygdaia, and a variety of sub-cortical structures (5~-52) . These brain regions are

involved in memory. learning, and cognition. Thus alterations in central cholinergic activity

often result in disruptions in the functioning of the neocortex such as disruptions in memory and

learning (26, 28, 32-36). Alzheimer's disease (AD) is the most common and severe Form of

central cholinergic impairment (53-55). This progressive neurodegenerative disorder is

characterized by deteriontions of memory, cognitive dysfunction. behavioural disturbances. and

deficits in activities of daily living (56-58).

Neuropathology of AD: the cholinergic hypothesis

In AD, neurochemical and neuropathological degenerative findings in central cholinergic

neurons are consistently reported (59-6 1 ). Neuroanatomically, AD is defined by neurofibri l lary

tangles and P-amyloid protein deposits, the characteristic brain lesions of this disease (1). There

is also a widespread decrease in the level of ACh in nerve terminais of the central cholinergic

system. This is due in part to a decrease in levels of ChAT and subsequent reduction in .\Ch

synthesis (7-9). In some brain regions, concentrations of ChAT are reduced by as much as SO-

90% (62). There are dramatic losses of neurons and synapses in many areas of the brain

including the basal forebrain, arnygdda, hippocampus, and cerebral cortex (5, 6) . The profound

loss of cholinergic innervation in the cerebral cortex Led to the cholinergic hypothesis of AD,

first proposed almost 20 years ago (1 1). This hypothesis is based on findings thai suggest that

central cholinergic pathways play a vital role in memory and cognitive function. In bi-ieE, the

evidence in support of the cholinergic hypothesis of AD is as follows: 1) cholinergic

neurotransmission is critical b r cognition in humans and animals (26. 29. 30, 33, 63-65); 2) the

blocking of cholinergic neurotransmission impairs cognition in humans and animais (29. 33. 34.

64. 65); 3) cholinergic markers are decreased in AD in a fashion that correlates with

neuropathology (66-69); and 4) patients with AD are particularly sensitive to choliner_ric

blockade (70-72).

The Cholinergie Hypothesis: Not the whole story in AD

Post-rnortern studies have demonstrated considerable heterogeneity in the cholinergic deficirs

seen in AD patients (52, 67, 73). As well, studies using cholinergic antagonists have shown that

these deficits alone do not fully account for al1 the cognitive and behavioural changes seen in AD

(-13, 71). Thus. although changes in the cholinergic system are reflected in cognition. other

pathological factors are at play. In fact, evidence shows disruptions in neurochemical marken in

the noradrenergic, serotonergic, doparninergic, and histaminergic systems in AD (7. 7 5 . 76).

[t is important to understand the contribution that changes in each of the various central

neurotransmitter systems makes to the overall clinical presentation of AD in an individual.

IdentiQing the heterogeneity of this disorder rnay be important for diagnosis and treatment; but.

at present, the understanding of the various subtypes of AD is limited. The development of a

peripheral cholinergic probe to assess centrai cholinergic activity rnay have important clinical

implications in tems of monitoring disease progression and treatment.

4 peripheral cholinergic probe as a mode1 for measuring central cholinergic ictivity

The physiologv of' the peripheral cholinergic systent of the iris

The pupil has exclusively autonomic innervation. Pupil diarneter is determined by the balance

between sympathetic and parasympathetic inputs to the two muscles of the iris. The sphincter

pupiiiae (constriction) is imervated by parasympadietic fibers onginating in the kdinger-

Westphal portion OF the nucleus of the oculomotor nerve (N III). These fiben synapse in the

ciliary ganglion and are distributed to the sphincter pupillae. The dilatator pupillae muscle

(dilatation) is innervated by long ciliary nerve branches of the nasociliary nerve branch of the

ophthalmic division of the tngeminal nerve (N VI), which cames sympathetic fibers that have

synapsed in the superior cervical ganglion (Figure 2). Clinically, cholinergic and antieholinergic

rye drops are used for examination. diagnosis. and treatment of various ocular conditions. These

topically applied drugs act on the sphincter pupillae muscle (77. 78). The response is mcdiatcd

by the rnuscarinic ml, mz, and rn, receptor subtypes (79, 80). The cholinergie antagonist

tropicamide, a muscarinic antagonist (8 1 ), dilates the pupil through paralysis of the sphincter

pupillae muscle. Pilocarpine, a muscarinic agonist (8 1 ), like Ach. directly stimulates the

sphincter pupillae causing rniosis (77, 78).

Previous m e of choIinergic and antîch olinergic eye drops: a diagnostic test for AD ?

Pupillary response to tropicarnide eye drops was initially proposed as a diagnostic test for AD

(23) as evidence began to accumulate suggesting that changes in the central cholinergic system

may be accompanied by changes in peripheral cholinergic systems (82-85). Early on. studies

revealed that individuals with Down's syndrome, a condition associated with reduced cholinergic

function, exhibited a greater increases in heart rate in response to atropine (a muscarinic

Figure 2 - Iiinervatioii of the Pupillary Muscles.

Adapted Troin Alder's Pliysiology of the Eye, O"' Edition (78).

antagonist) than healthy controls (84). These patients also had a significantly greater mydriatic

pupillary response to atropine and tropicarnide than controls (83, 85). Recently. pathology of the

Edinjer-Westphal nucleus has also been established in AD (56). One paper went on to suggest

that a tropicamide eye &op test may be able to distinguish individuals with AD from those with

other foms of dementia (83), as a similar pattern of brain lesions and related dementia is prescnt

in adults with Down's syndrome and AD (10). These observations Ied Scinto et al. (33) to

investigate the possibility that pupillary challenge with anticholinergic eye drops may serve as a

diagnostic test for AD. Their result suggested that subjects with AD did in fact display a

hypersensitive mydnatic response to tropicamide when cornpared to age-matched control

subjects. At minute 29 post eye drops, on average, there was a 73.1% change in pupii diameter

of AD patients compared with a 5% change for healthy subjects (p=0.009). However. numerous

studies were conducted attempting to replicate these results, but were unable to achieve the same

separation between the hvo groups. These other studies observed a similar degree of pupil

mydriasis in the AD patients (19.7% to 22.5%). but a considerably larger response in the control

groups ( 19.1% to 23.4%) (24.87-89).

Other studies exarnined the potential of pupillary response to pilocarpine as a diagnostic test for

AD. One such study showed that the miotic response of AD patients to pilocarpine (45 minutes

post eye drops) was significantly greater than that of age-matched control subjects (p<0.001)

(90). This result was consistent with the theory that deficits in the peripheral cholinergic systrm

occur in paralle1 with central cholinergic decline. A compensatory upregulation of cholinergic

receptors would be expected to occur in response to the decreased in fiction. which would

result in the hypersensitive response observed with a challenge by an agonist. In contrast ro the

tropicamide eyedrop literahue, the five studies using pilocarpine eye drops al1 found a

hypersensitive miotic response in AD patients compared to controls (90-94).

A Tropicamide-Pilocarpine challenge test (TR OPIL-Cg

Our group proposed that pupillary response to cholinergic challenge rnay be an index of central

cholinergic function and not a marker for the presence or absence or AD (3). The conRicting

results from the diagnostic studies may have been due to heterogeneity in central cholinergic

function within disease States or differences due to normal aging. We evaluated the ability of a

tropicamide-pilocarpine challenge test (TROPLL-CT) to detect differences in central cholinergic

function in healthy young and elderly subjects. TROPIL-CT involved inducing mydriasis with

tropicamide and then opposing the effect with pilocarpine. The procedure was first optimized by

showing that there was not a significant difference in response behveen young and clderly

subjects to tropicamide alone. The rationale of using tropicamide was to induce rnydriasis.

which would presurnably lead to a more observable miotic effect by pilocarpine and prevent a

floor effect of pupil contraction. The results fiom TROPIL-CT showed that there was a

significantly greater miotic response to pilocarpine in elderly subjects (59-80 yn) at 85. 125.

165. and 215 minutes post pilocarpine, when compared to the young subjects (15-40 yrs)

(~~0 .05 ) . For example, at 21 5 minutes post-pilocarpine administration, the p lacebo-corrected

net decrease in pupil diarneter fiom baseline measurements was 8.8% k 4.9% in the young and

23.9% * 2.4% in the elderly (p40.05). Thus, the findings with tropicamide and pilocarpine were

consistent with the previous studies and provided fùrther evidence that pilocarpine detects

differences in central cholinergic activity better than tropicamide.

The TROPIL study also examined the effect of a low dose of scopolamine, a centrally acting

cholinergic antagonist, in the elderly and the young groups to venQ that the elderly group did

indeed have decreased central cholinergic function cornpared to the young. Previous studies had

demonstrated that scopolamine was able to reflect differences in central cholinergic function, as

individuals with central cholinergic deficits showed greater cognitive impairment resulting from

treatment (34-36). The elderly subjects in the study had larger central cholinergic impairment

relative to the younp subjects. The elderly expenenced a greater degree of cognitive impairment

in word recall, at 60, 90, and 120 minutes afier receiving scopolamine, compared to the young

(pe0.05). However, no linear correlation was found between the centrai and peripheral

resporisés (PO. i 3, p-0.60). However, in the eidtrly, three subjecrs haci iciiosyncraric responses

and appeared to be outliers: one had an early and accelerated response to pilocarpine; the second

had an exaggerated response to scopolamine (floor effect); and the third had no response to

scopolamine. Thus, the responses of 2 patients may be an anifact generated by using

scopolamine as a mode1 and the third is unexplained. When these subjects were removcd from

the analysis. there was a significant correlation between the responses ( ~ 0 . 6 6 . p<O.Oj)

21 Pilucarpirie challenge test (PIL-CT)

The TROPIL-CT probe was designed to capitalize on the homeostatic cholinergic control of

pupil size in such a way that changes in size would be maximized. It was expected that initial

dilation of the pupil with tropicamide would allow for preater miotic changes due to

administration of pilocarpine. While the magnitude of the change was increased. the length of

tirne required for miosis was lengthened by the initial dilation with tropicarnide. In an attempt to

simplify the procedure, particularly for the use in elderly and AD patients, the rnethodology of

TROPIL-CT was altered to involve only pilocarpine administration. The feasibility of using a

pilocarpine challenge test (FIL-CT) to mesure central cholinergic f ic t ion was evaluated in 4 of

the healthy elderly volunteers from the previous TROPIL-CT study as individual results from

both methods were compared. The maximal decrease in pupil size for PL-CT (45 min post-

eyedrops) was 29.6% ir 4.03% compared to 29.7% + 3.06% for TROPIL-CT (85 min post-

eyedrops). Thus, the response to pilocarpine alone was faster and of similar magnitude. When

comparing individual responses, the net difference between the two procedures was 0.95" +

0.50%. The result of a paired t-test indicated that there was not a significant difference in the

miotic response to PIL-CT compared to TROPIL-CT for the elderiy (n.s.) (25) .

Crirrent Dirmiorrs

These preliminary findings appear to support the proposal that differences in central cholinergic

functioning in the elderly can be detected using pupillary responsiveness to pilocarpine. .A

logical extension of this work is to examine the utility of using pupillary responses to mrasure

greater deficits in central cholinergic function in disorders such as AD. Furthemore. this probe

rnay have clinicai applications in idenriQing subtypes of AD or as a predictor of treatrnent

response to cholinergic therapy.

Cholinergie Replacement Therapy for AD

The evidence supporting the cholinergic hypothesis of AD suggests that cholinergic replacement

rnay improve cogition (95. 96). Similar to the concept OF treating Parkinson's disease with

doparninergic agents, initial approaches to treating AD were to enhance cholinergic function

(56). Four strategies to augment cholinergic transmission are available. The first involves

administration of precurson to ACh synthesis (e.g. lecithin or choline). but clinical results have

been disappointing due to the very high dosages required to overcome peripheral breakdown and

to achieve reasonable brain concentrations (97-104). The second approach involves

administenng cholinesterase inhibitors (ChEIs) to block enzyme activity, thus increasing the

availability of ACh by reducing its degradation (1416). This approach has been the most

shidied and ChEIs have proven to be effective (15, 19-22, 105, 106). The third and fonh

strategies involve the use of partially selective muscarinic or nicotinic receptor agonists and

indirect cholinergic receptor agonists to augment cholinergic function, both of which are

currently in the early developmental stages (107, 108).

Cholinesterase inhibitors

Currently, only one ChEI, donepezil (AriceptB), has been approved by the Health Protection

Branch of Canada for the treatment of AD. The United States Food & Drug Administration (CS

FDA) has also approved tacrine (cognexB) along with donepezil. A number of ChEIs are

expected to join the market in the next few years as trials for several other agents (ce. C

rivastigmine, galantamine) are well undenvay ( 13, 18. 109).

The results from the meta-analysis of the efficacy and safety of oral ChEI treatment in ,AD

(please see Study 2) demonstrated that the proportion of AD patients showing a signifiant

improvement in cognition was 31% [95% CI: 24%, 38%]. However, this was only 1296 [9j06

CI: 7%, 17%] better than placebo. It has been proposed that variability in treatment response

may be contingent on the integrity of the cholinergic system, since these medications can only

act by preventing the breakdown of pre-existing acetylcholine (1 10). Therefore, AD patients

with advanced disease progression and greater cholinergic deficits rnay be unable to respond

significantly to treatment (12, 11 1). Individuals with similar clinical presentations may have

different levels of endogenous cholinergic activi ty, and thus respond di fferently to treatment. .A

non-invasive, simple, sensitive, and reliable measure of central cholinergic function may identify

subgroups of patients based on disease progression or disease subtype that rnay be better suited

for ChEI treatment (1 12).

Study 2 dernonstrated that many AD patients might undergo costly treatments with minimal

benefit. Currently, making the distinction between treatment responders and non-responders

oRen requires months, a time period during which a substantial irreversible decline in cognitive

lunctioning otten occurs. As more treatment options are on the horizon, AD patients not suited

for ChEI phamiacotherapy could be placed on other non-cholinergie therapies instead. As the

popuiation continues to age, it is projecred that the prevaience of AD will m e three fold to -600

000 in Canada by 2021 (2). This is a major cause for concem, which has lead to research

involving the search for a simple test for early and accurate diagnosis and predictors of treatment

response to medications for AD.

AD patients will show an increase in peripheral cholinergic function (e-g. PIL-CT response )

compared to healthy elderly controls due to deficits in central cholinergic Functioning resulting

from disease pathology.

~ ~ [ E T H O D S FOR STUDY 1.1

Subjects

.4D patients with a diagnosis of probable Alzheimer's type dementia according to the Diagnostic

Statistical Manual of Mental Disorders, Founh Edition (DSM-IV) criteria ( 1 13) and the National

Institute of Neurological and Communicative Disorders and Stroke - Alzheimer's Disease and

Related Disorden Association (NINCDS-ADRDA) cntena ( 1 14) were recmited from an

outpatient population drawn fiom the genatric psychiatry clinic at Sunnybrook and Women's

College Health Sciences Centre (S&WCHSC). The files of the Community Psychogeriatnc

Service for the Elderly (CPSE) were also examined, advertisements were placed in local

newspapen. and farnily physicians in the comrnunity were consulted for potential study

candidates. Elderly control subjects were recruited From a previous study conducted iit

S&WCHSC.

Inclusion Criteria:

AD patients were:

2 55 years, and met critena for probable AD according to the NTNCDS-DRDA and the

DSM-IV cnteria for degenerative dementia

O had an Mini Mental State Examination (MMSE) score in the range of 1 1 to 24

had not have received a ChEI during the past 4 months

Elderly control subjects had an MMSE score no less than 28.

A signed statement of consent was also required by the subject or. in the case of the AD patients.

their legal representative (Appendix A).

Exclusion Cr iteria :

ophthalrnologic diseases (e-g. glaucorna)

neurological disorders which would interfere with diagnosis of probable AD. such as

Parkinson's disease or stroke

documented hisrory of psychiatric illness prior to diagnosis of dementia. such as

schizophrenia, bipolar and major depressive disorder

0 rnrdications with known cholinergie activity

concomitant medications remained unchanged during the study. Subjects refrained from

alcohol, smoking, and food and beverages containing caffeine for at least 12 hours before and

during the test session and antihistamines Cor the entire duration of the study. This study was

approved by the Research Ethics Board of S&WCHSC.

S tudy Design

Subjects attended one study session. Upon arrival, subjects and their legal representatives were

infonned as to the nature and purpose of the study and infonned written consent was obtainrd.

A dit lamp assesment was also performed by an ophthalmologist (Dr. Catherine Birt) to screen

for ophthalmologic diseases that would have excluded a patient from the study. An MMSE was

also performed to detemine if the subject's score was within the appropriate range for inclusion.

This was followed by a pilocarpine challenge test (PL-CT) of the subjects' peripheral

cholinergic system of the iris, as descnbed below.

Study medications

To measure peripheral pupillary responsiveness to PIL-CT, a pilocarpine eye drop (20 pl. O. 1 % )

was administered to one eye, while a placebo eye drop (20 pl, saline) was administered to the

other eye, in a randomized order. The onset of the observed miosis begins approximately 15

minutes afler instillation, with a peak at about 45 minutes (1 15). Pupillary diarneter uVas

measured at the following times: twice at baseline (8 minutes and 5 minutes before eyedrops). 3

minutes post eyedrops, and peak response times of 20, 45, and 60 minutes. Pilocarpine is a

muscarinic cholinergic agonist, which causes a miotic response in the pupil and is widely used as

an effective agent for controlling intraocular pressure and preventing blindness in jlaucoma

patients (1 16). Pilocarpine solutions were prepared by the S&WCHSC Pharmacy using a sterile

ophthalrnic irrigating solution that contained Tears ~ a t u r a l e ~ (Alcon. Inc.).

Measurement of Pupil Diameter

Pupil diarneter was rneasured using a previously validated unilateral computerized infrared

pupillometer (1 17). This apparatus consisted of four infkared diodes that illuminateci the eye, an

infrared sensitive video canera (Panasonic GP-MF502), and a computer-controlied display

television to focus images of the eye. Each measurement of pupil diameter was determined from

the maximal diarneter value out of 40 cornputer-generated diameter measurements of captured

video images processed to determine the edge of the pupil. Measurements were taken in a quiet

room, with a standard illumination of 750 lux. Subjects were seated in an armchair and. when

pupillary measurements were taken, placed his or her head on a chin-rest connected to the

adjustable camera stand. Subjects were also asked to locus on a target placed at a standard

distance behind the pupillorneter (Figure 3A).

A second pupillometer was also employed to measure pupil diameter. This bilateral apparatus

consisted of 12 infrared diodes that illuminated each eye, two infrared sensitive video cameras

(Fujinon 609478) mounted in parallel, and a computenzed display. Pupil images were

simultaneously focused on a computer screen and series of 3 pictures, taken in np id succession.

were captured and stored. ARer each session was completed, each set of images was processd

by computer to determine the mean value of the 3 pupil diameter measuremrnts at each timr

point. Pupil measurements were taken in the same room as descnbed above. .A special mirror

allowing infrared light to pass through it was placed between the patient's eyes and the video

cameras. The mirror was placed at an angle and subjects were asked to focus on the retlected

image of a light bulb in the centre of mirror (Figure 3B).

Instiilatioo of eye drops

The procedures for the instillation of the eyedrops were carefully controlled to reduce factors that

decrease pilocarpine concentration at the source. Subjects were in a supine position. with his or

her head tilted back at a standardized angle. With the subject's eye open, the investigator Formed

a pouch by gently lifting the lower eyelid. The subject was instnicted to look up (towards their

own eyebrows) to minimize the blink reflex. The subject was also instnicted to apply pressure to

Figure 3A - Schernatic diagram of unilaterai pupillometer

Figure 3B - Schematic diagram of bilateral pupillorneter

his or her own nasolacrimal duct as the drop was instilled into the pouch formed by the lified

eyelid. The use of a micropipette allowed for accurate administration of a small volume drop (20

pl), which rninimized tearing. The subject was then instructed to look down slowly. maxirnizing

comeal contact with the instilled drop. Afler 30 seconds, the subject was instructed to close their

eyes For 2 minutes. Throughout, the subject continued to apply pressure to their nasolacrimal

duct to minirnize the arnount of dnig escaping to the systemic circulation. Aller 2 minutes. the

subjeci was returned to an upright position and pupil size was rneasured as outlined above.

Safeîy

Pilocarpine has been approved by the Health Protection Branch of Canada as an topical rniotic

and for the treatrnent of ocular hypertension and chronic open angle glaucoma. The most

frequently observed transient and mild adverse effects of pilocarpine are temporary pupil

constiction. reduction in intraocular pressure, ciliary spasms. loss of visual acuity, and myopia.

Possible systemic cholinergie adverse effects include increased sweating ( 1 15). The fiequency

of systemic advene effects was minimized by applying pressure to the nasolacrimal duct during

drop instillation and by using subclinical doses of pilocarpine.

Data Analysis

Data were analyzed using the Statistical Program for the Social Sciences (SPSS) for Windows.

version 10.0. Percent decreases in pupil diameter were expressed at each time point as pupil

diarneter for pilocarpine minus the adjusted pupil diarneter for placebo divided by the adjusted

pupil diameter for placebo at that time point multiplied by 100%.

For example at time point 5:

% decrease in pupil - - &&

diarneter at time point 5 b h d j .u 100%,

where pupil diameter for pilocarpine at time point 5 is AS and the placebo adjusted pupil

diarneter at time point 5 is BSAdj.

The adjusted pupil dimeter for placebo took inio account the percentage change in the diamrter

of the placebo eye fiom baseline.

For example at time point 5:

B5:\d1 = B5 - [Bs .Y (BB - B s ) ~ B ] ,

where pupil diameter for placebo at time point 5 is B5 and pupil diameter at baseline is Bu.

Thus, (BB - Bc)/Be is the percent change in placebo pupil diameter at time point 5 from baseline.

which was multiplied by pupil diarneter at time point 5 (Bs) to give the absolutc change from

baseline for placebo. This value was then subtracted fiom the pupil diameter at time point 5 to

detemine the placebo adjusted pupil diameter for time point 5 (B5hdl).

RESULTS FOR STUDY 1 .l

Subjects

4 healthy elderly (mean age 67.5 1 3 . 0 yrs, range 59 to 72, MMSE 29.0 t 0.4) and 6 AD patients

-1 (iiiéiiii ilse 77.0 F 2.9 yrs, raiige 65 io 67, MMSE 21.3 i i.9) w r e reçrui~ed. i ne mew ages

benveen the AD patients and the healthy elderly controls were not significantly different

(p=0.06). The difference in MMSE was statistically significant (p=0.005). Other patient

characteristics such as weight and height were not significantly different (n.s.). Al1 siibjects were

of European descent except for 1 AD patient who was of Guyanese descent (subject 2). ;\Il

subjects had blue eyes except for one elderly subject (subject 4) and one AD patient (subject 2 )

who had brown eyes (Table 1).

Pilocarpine effect on pupil size

The time course of changes in pupil diameter following pilocarpine administration in the hcalthy

elderly and AD patients is presented in Figure 4. The miotic effect of pilocarpine occurred faster

in the AD group. There was a significant difference in pupillary response to pilocarpine at the 3-

minute time point (p=0.05). There was no difference between the two groups at the 20-minute

time point or beyond (m.). Ader 20 minutes, the healthy elderly group had a larger response to

the pilocarpine challenge than the AD patients did. The peak decrease in pupil size for the .U

goup was 28.6 t 7.3 % and was seen at the 45-minute tirne point. For the heaithy elderly. the

peak e&ct was 33 -5 t 6.0% and was observed at the 60-minute time point.

Table 1 - Characteristics of elderly and AD subjects --

Subject Age, Gender Iris Concomitant rnedicatioiis Yr Colour

- -

Elderly

1 72 M Blue none 2 71 M Blue cilazapril 2.5 mg qd 3 59 F Blue cilazapril 2.5 mg qd, giblenclan~ide 2.5 mg qd, ASA 325 nig qd 4 68 M Brown none

Mean 67.5 (SEM) (3.0)

78 F B 1 ue calcium carbonate 1000 mg qd, 75 M Browii felodipine 10 nig qd, labetülol HCI 100 nig hid, risperidone 1 nig qd, lorazepam 0.5 mg hs

Pr'-' 87 F Blue verapamil 120111 mg tid, levothyroxinc Na 0.05 nig od, calciuiii carbonate 500 nig bid,

alendronate Na I O mg qd 78 M Blue nifedipine 60 nig qd, EC-ASA 325 mg q2d 65 M Bluc none 79 M Blue diltiazem 300 mg qd, hydroclilorotliiazide 25 mg qd

Mean 77.0 (SEM) (2.9)

auyasea moq JaJameia adnd a! aseamaa %

There is a correlation between treatment response to the cholinesterase inhibitor donepezil and

peripheral cholinergie hnc tion.

METHODS FOR STUDY 1.2

Subjects

AD patients completing the first phase of the study were informed as to the nature and purpose

of the second study and those who provided informed written consent proceeded to the ChEI

treatment phase of the study (Appendix B). Additional patient cntena were:

E~cl~ision Criteria:

hypertension, defined as blood pressure greatrr than or equal to 160 mmHg systolic and 100

mmHg diasto lic pressure

other severe cardiovascular disease, such as myocardial infarct. or severe angina psctoris

within the last year which could jeopardize ability to complete the triai

a se ime disorders, asthma, or contraindications to treatment with donepezil

Study Design

Patients attended 1 study session per month for 4 months, for a total study length of 12 weeks.

Al1 patients received active treatment with the ChEI donepezil. Patients received 5-mg

donepeziYday for 4 weeks, which was increased to IO-mg/day for 8 weeks, if tolerated. The

baseline assessrnent was conducted in conjunction with the session conducted for the tint pan of

the study. During the baseline and final visit, al1 subjects undenvent the following cognitive and

fùnctional testing: the ADAS-cog (Alzheimer's Disease Assessment Scale, cognitive subscale)

(1 18), the MMSE (Mini-Mental State Examination) (1 19), the DAD (Activities of daily living:

Disability Assessment for Dementia) ( i 20), and the CIBIS-plus (+) (Clinician's Interview-Based

Impression Disease Seventy with Caregiver Lnput) (12 1 ). These tests were followed by

administration of the PIL-CT, as previously descnbed. During visits 2 and 3, oniy the . M X S -

cog and PIL-CT were performed. During the first visit. screening procedures performed

included an rlectrocardiograrn (ECG). In addition, a blood sample ( I 5-ml) was collected at

screening for routine analysis (hematology/biochemistry) (Figure 5 ) .

Baseline Week 3 Week 8 Week 12 Visit (Visit 2) (Visit 3) (Visit 4)

Screening Procedures:

Outcome Measures Performed:

Treatrnent Design:

Blood Samples Collected for:

Ophthalmologic ECG

ADAS- CO^ MMSE DAD CIBIS+ PIL-CT

5-mg DNZ

routine analysis

PIL-CT

IO-mg DNZ

PIL-CT

1 O-mg DNZ

ADAS-cos MMSE DAD ClBlS- PIL-CT

Figure 5 - Summary of study design for Study 1.2.

Treatment Response Measurements

Our prirnary outcome measures included the baseline PIL-CT score and the post treatment

change in ADAS-cog score from baseline.

The ADAS-cog is a sensitive and reliable neuropsychological scale used by the US FDA as one

of the standard instruments for evaluating cognitive enhancer dmgs seeking approval ( 121). This

multi-item scale examines selected elements of memory, orientation. attention. reasoning.

language, and praxis (123). Scores range From O to 70. with a higher score characterizing a

greater degree of cognitive impairment. A positive treatment response is generally defined as a

4-point irnprovement in the ADAS-cog from baseline (1 5 ,2 1,22, 124).

Our secondary outcome measures included the MMSE. the DAD, and the CIBIS+.

The MMSE is a simplified scale that focuses on the cognitive aspects of mental functions. This

scale assesses orientation, memory, attention, and praxis with scores ranging [rom O to 30. For

the MMSE. a lower score indicates a greater degree of impairment (1 19).

The DAD is a scale used to measure the functional impairment as indicated by a decline in

performance of activities of daily living. This scale is administered through an interview with

the caregiver, and examines items such as personal hygiene, dressing, eating, going on an outing.

finance and correspondence, medications, leisure, and housework. Scores are expressed as a

percentage of applicable items successfully initiated, planned, or performed, with a lower score

characterizhg a greater degree of impairment (120).

The CIBIS+ is a global rating scale that provides an index of clinical impression of change that

cannot be obtained fiom quantitative assessrnent measures. This scale is performed through an

interview with both the patient and the caregiver and identifies four major categories for

evaluation: general, mental/cognitive state, behaviour, and activities of daily living. These four

catesories are further subdivided into domains, which are assessed by the use of probes including

the MMSE and DAD described above. A 7-point Likert-type scale is used to assess change from

baseline, with intervals ranging fiom I ("very much irnproved") to 7 ("very rnuch worse"). with

4 indicating no change since baseline ( 12 1).

Safeîy

Donepezil has been approved by the Health Protection Branch of Canada for the treatment of

AD. In a randomized, double-blind, placebo controlled trial of 5-mg donepezi liday invo king

14 1 patients, the incidence of treatment-emergent adverse events was comparable to t hat

observed with placebo (15). The most frequently observed adverse effects were nausea.

vomiting, diarrhea, diuiness, gastric upset, and constipation. These rvents are generaily mild

and transient. subsiding within a few days of beginning treatment. A larger incidence o l adverse

effects is observed at a 10-mg/day dose, although the frequency of adverse effects is less so long

as patients receive 5-mg/day for 3 weeks pnor to dosage increase (124). In an open-label study.

patients undergohg a six-week titration to 10-mg/day showed a sirnilar incidence of adverse

effects as 5-rng/day and placebo groups (14). In ou study, patients received 5-rnyday of

donepezil for 4-weeks before the dosage was increased to IO-rnglday, thereby minimizing the

likelihood of adverse effects.

Data Analysis

.4ccording to sample size calculations, 26 subjects should have been recruited to complete the

study. As a dropout rate of approximately 20% was expected, the target nurnber for recruitmrnt

was 32 subjects in order to have 26 subjects complete the trial. The actual number of subjects

recruited was 6. 1 subject dropped out of the study. thus 5 sub!ects completed the trial. The

small sample size piaced limitations on the statistical analyses that could be performed. Because

of this, the results were comprised of mainly descriptive trends. However, the followinp

statistical analyses were performed:

To test the hypothesis that treatment response to ChEIs c m be predicted From periphrral

cholinergie responsiveness, correlation analyses were performed on baseline (pre-treatment) PIL-

CT responses and the end of treatment cognitive and functional outcome measures ( e g ADAS-

cog, MMSE, and DAD). Correlational analyses were also performed to detemine whethrr

cognitive state (i.e. ADAS-cog score) was related to PIL-CT response over the course of

treatment with donepezil for the group and individually.

RESULTS FOR STUDY 1.2

Subjects

16 patients were screened for entrance into the study. 8 failed to meet inclusion/exclusion

criteria. 8 patients were enrolled in the study; however, 2 subjects never begm the rtudy due !o

withdrawal of consent for one and hospitalization for the other. Thus. 6 AD patients cornpletrd

Study 1.1 and were recruited for Study 1.2 (Table 1). 1 subject (subject 1) discontinued the

study treatment after 2 days due to adverse events and could not be included in the analyses.

Subject 5 did not tolerate 10-mg/day of donepezil and the dose was reduced to 5-myday for

weeks 9- 12 (Figure 6) .

- - -

( 16 patients screened 1

8 excluded w= 8 patients entered into study

3 discontinued 1 adverse event 1 withdrawal of consent 1 hospitalization

1 Figure 6 - Trial Profile for Study 1.2

Efficacy Assessrnent

A11 outcome measures are summarized in Table 2.

PIL-CT. At baseline. the peak decrease in pupil diarneter for the AD patients was 32.6 + S.Oo~o.

AFter 12 weeks of treatment with donepezii, a 13.3 i 7.3% decrease in responsiveness to PI L-CT

was observed (Figure 7A). 4 of 5 patients showed a decrease in pilocarpine response. whereas

the other subject had an even greater miotic response.

ADAS-cog. AD patients had a mean ADAS-cog score of 23.3 f 3.8 at baseline. After 12 weeks

of donepezil treatment, the group showed a 4.8 f 1.4-point improvernent in ADAS-CO, u score

(Figure 7B). There was an improvement in ADAS-cog score From baseline in al1 5 participants

in the study. 3 of the 5 subjects had an improvement of 4 points or greater (sigificant

responders), inciuding 2 subjects improving by 7 points or greater (Table 2 ) . The other 2

subjects had a 1-point improvement in their ADAS-cog scores.

Seco~idqv Oittcome Meastires

MMSE. At 12 weeks there was a 2.4 k 0.5-point improvernent in MMSE score From the baseline

mean of 19.8 +_ 1.8. Al1 5 subjects showed an improvement in their MMSE score. Z subjects had

4-point and 3-point improvement in MMSE score, respectively. 2 subjects improved by 2 points.

The remaining subject's MMSE score improved by 1 point.

Baseline Week 12

Fiaure 7A w - Cornparison of mean baseline and post-donepezil PIL-CT response.

Fiaure 7B

Baseline

- Com~arison of mean baseline and ~ost-done~ezil ADAS-coa score

Week 12

DAD. At baseline. the group scored 70.0 t 14.6% on the DAD. M e r 12 weeks of donepezil

therapy. there was a 2.2 f 5.7% increase in activities of daily living as measured by the DAD. 4

of 5 subjects demonstrated an improvement in their DAD scores. whereas 1 subject had a

substantial decrease in function.

CIBIS+. CLBIS+ evaluations were only obtained for subject 4 to subject 6. At baseline. 2

subjects were received a rating of mildly ill. The other was rated as moderately ill. At 1 2 weeks.

1 subject was rated a worsened, 1 was rated as no change, and the remaining subject was rated as

improved over the course of treatment.

PIL-CT as a predictor of ChEI response

There was not a significant correlation between baseline PIL-CT response and improvement in

ADAS-cog score in the 5 subjects (r=0.22, p=0.73) (Figure 8A). The mean baselint pupillary

response of the 3 significant responden (2 4-point improvement in ADAS-cog) to donepezil

therapy was -29.8 + 13.3% compared to -36.9 + 9.0% for the non-responders (Figure 9).

Similar analyses were performed with MMSE, DAD, and CIBIS+ scores. For the MMSE. a 3-

point improvement was considered significant. For the DAD a 2 9% irnpmvernrnt was

considered significant. For the CIBIS+, the rating at the end of week 12 was taken as the score

(e.g. improvement, no change, wonening).

There was a strong correlation between baseline PL-CT response and improvement in MMSE

score, but the correlation was not statistically significant (r=0.73, p=0.17) (Figure 8B). The

rnean baseline pupillary response of MMSE responders was -26.1 t 19.8% compared to -37.0 k

7.7% for the non-responders (Figure 9).

lrnprovement in ADAS-cog

-2 5 -1 5 -5 5 15 % lmprovement in DAD

lmprovement in MMSE

Worsening No Change lmprovement lrnprovement on CIBIS+

Figure 8 - Summary of correlations between baseline PIL-CT response and treatment response to donepezil according to various outcome measures

MMSE

OResponders [al Non-Responders

DAD

Figure 9 - Summary of PIL-CT responses for responders and non-responders to donepezil treatrnent according to

various treatment outcome measures

There was not a significant correlation between baseline PIL-CT response and change in D m

score (r=0.02. p=0.98) (Figure 8C). The mean baseline pupillary response of DAD responders

was -27.6 i 11.5% compared to 10.2 k 12.3% for non-responders (Figure 9).

.A correlation could not be calculated for CIBIS+ scores because of the sarnple s i x (n=3). but

baseline pupillary response was plotted versus final CIBIS+ with improvement increasing from

lett to right to determine if' a trend was present (Figure SD). The mean baseline pupillary

response of the CIBIS+ responders was -45.8% compared to -17.1 I 10.896 for the non-

responders (Figure 9).

PIL-CT as a rneasure of disease state

i\DAS-cog score was correlated with PIL-CT response at the sarne session for al1 subjects

(Figure IOA) and for each subject individually (Figures 1OB - IOF). A significant correlation

was not seen in the group as a whole, and though trends were seen in 1 of 5 subjects oniy one

correlation was statistically significant (subject 2; r-0.999, pc0.001) (Figure 1 OB).

ADAS-cog score ADAS-cog score

C 0

Subject 3 D o .

Subject 4

O 10 20 30

ADAS-cog score ADAS-cog score

E o . Subject 5

F O Subject 6

O 5 1 O 15 20

ADAS-cog score ADASlcog score

Figure 10 - Summary of correlations between PIL-CT response and absolute ADAS-cog score. (A) Correlation between al1 subjects, (B) to (F) lndividual correlations for subject 2 to 6.

This study is a Iogical extension of previous work that demonstrated that a tropicamids-

pilocarpine challenge test (TROPIL-CT) of the peripheral cholinergic system of the iris may

detect differences in central cholinergic function in healthy young and elderly subjects (25) . .A

peripheral probe of the central cholinergic system has considerable potential as clinical tool ro

diagnose, monitor, and treat disease States affecting the cholinergic system such as .AD. 1Ve

continued this line of research by conducting a hvo-part study (Study 1 ).

The discussion must begin with an explanation for the limited number of subjects involved with

both components of this study and the implications that a small sample size has in ternis of

interpreting the results of this study. Clinical research is notorious for difficulties with recruiting

subjects and recruitment for this study was fraught with its own unique challenges. The major

obstacle that presented itself for the prediction study (Study 1.2) was that the study rnedication.

donepezil, was approved by the Ontario Drug Benefits Formulary (O.D.B.) on a limitrd use

based in conjunction with the manufacturer (Pfizer Canada Inc.) shonly after recniitment for the

study began. As compensation for participating in the study, subjects were initially offered a 6

months' supply of donepezil, valued at approximately $8 19 (1 82 caps @ S4.5Okap). ARer the

drug's approval, the only compensation available was the close monitoring associated with

participation in the study. Udortunately, this benefit doubled as a deterrent to potential

participants who did not want the hassle of multiple hospital visits associated with their ChEI

treatment. The addition of donepezil to the O.D.B. also reduced the potential patient population

as many AD patients, to whom the high cost of the medication had previously prohibited them

From ChEI therapy, were subsequently treated soon afler approval.

The Cornrnunity Psychogeriatric Service for the Elderly (CPSE) at Sunnybrook & Women's

College Health Sciences Centre (S&WCHSC) was also consulted in the search for potential

subjects. The CPSE sees 175 new cases/year (almost 100% dernentia) and actively follows

approxirnately 250 cases. An examination of the patient files revealed that a large portion of the

cases had a history of psychiatrie illness (mostly depression or treatment with anti-depressants).

A list of 23 potential subjects was brought to the attention of the CPSE, and al1 but 7 subjecrs

were deemed ineligible or unlikely to participate. Two subjects were contacted by the CPSE and

did not wish to participate in the study. Four patients lived in nursing homes and suitable

arrangements to bring them to the hospital could not be made. The remaining patient was lost ro

follow up.

Two advertisements were placed in local magazines/newspapers. Unfortunately. only 7 inquiries

were made to the ads. In addition, the majority of callers were not caregivers of AD patients. but

individuals concemed with minor lapses of forgethlness expenenced by themselves or by

someone they knew. Due to the high cost and low return associated with this mode of

recmitment, further attempts to advertise in newspaperdmagazines were not made.

Attempts were also made to enlist the support the Departments of Geriatric Psychiatry. Geriatric

Medicine, Farnily Medicine, and other researchen conducting studies in dementia at

S&WCHSC. However, despite frequent reminders and visits to individual physicians. refernls

were not forthcoming. Presumably, as a tertiary care facility, the vast majority of potential

patients arriving at our site were not eligible for the study as they had complicated diagnoses,

were well advanced in the disease state, were already taking donepezil, and/or had already

attempted a trial with the ChEL

It was postulated that the ideal population for the study was newly diagnosed AD patients.

Again, as a tertiary care facility, S&WCHSC did not appear to be the best location to seek out

these individuals, as most new cases of AD are diagnosed by general practitioners in the

community and patients are immediately started on a trial with a donepezii. As a result. a

community research group was formed consisting of 20 physicians with clinics in the community

and who were cross-appointed to S&WCHSC or affiliated as courtesy staff. Members of the

research group received monthly newsletters describing ongoing studies and some physicians

were visited personally. Unfortunately, this group was not formed until the latter stages of this

project and the potential for recniitment From this source was not fully explored. The

community group provided one referral. but the patient was hospitalized and could not

participate in the study.

The level of recruitment for this study fell far below the targeted sarnple size. These results

underscore the challenges associated with carrying out clinical research at the present time. This

has far-reaching ramifications For the feasibility of these types of studies in academia.

particularly when snict inclusion/exclusion criteria are necessary. This study required .AD

patients who had yet to try donepezil, the first line and only ireatment for their disease.

Recruitment of subjects at S&WCHSC between lanuary 1999 and May 2000 resulted in the 6

AD patients included in the study. The cornmunity research group was not formed until January

2000. Expansion to a multi-centred trial was limited by the procedures employed in the study.

In pupillometry, differences in extemal conditions (e.g. lighting) may have a significant effect on

pupil size. Thus. it was important to maintain consistent testing conditions, as results between

subjects were being compared. The use of multiple testing sites would Further add to the

vanability already associated with the procedure. In addition, even if standardized sites could be

established, pupillometers are not commonly available and not readily portable. Taken together.

these challenges limit the number of AD patients that were recruited into the study.

The healthy elderly patients participating in Study 1.1 had previously participated in the project

comparing ocular responses between young and elderly subjects to TROPIL-CT ( 2 5 ) . Four

subjects From the TROPIL study participated in the prelhinary study evaluating the feasibility

of using pilocarpine alone (PIL-CT) to assess penpheral cholinergie responsiveness. These

results were used as the basis for cornparison with the rneasurements obtained from the AD

patients. The recniitment of additional healthy elderly patients was planned but was not

followed up, as the number of subjects in the AD proup remaineci low.

The statistical implications of the small sample size in this study are also an imponant issue to

address. This study was not able to detect a difference in pupillary response to pilocarpine

between healthy elderly controls and AD patients; however, the sample size did not have rhe

statistical power to reject the nul1 hypothesis that there was a difference behveen the two groups.

A post hoc power analysis revealed a 7% chance of yielding a significant effect such that the nul1

hypothesis, that the means of the two group were equal, could be rejected. A 95% confidence

interval indicated that the maximum response of the elderly controls could be a large as 27.7Ou

greater than or 17.9% less than that of the AD patients. The fact that a greater response of -Mo0

was seen in the elderly group compared to the AD group could no more be attributed to the

disease than to chance. A post hoc sample size calculation determined that 32 subjects would be

required in each group (n=64), assuming simîlar standard deviations in PIL-CT response For both

goups as seen in this study, to detect a 10% difference between healthy elderly controls and AD

patients, with a significance level of 0.05 and a statistical power of 0.8. Likewise, though no

correlation could be found between treatrnent response and PIL-CT response, the sample size

was too small to reject the possibility that a relationship exists between central and peripheral

cholinergie responses. A post hoc power analysis revealed that for this study, conducted in 5

patients, the probability that a relationship behveen ADAS-cog score and PIL-CT (with a

standard deviation of 3.36 for AD AS-cog scores and 18.6 for PL-CT scores) would be detectrd

was 14 percent (1-B), assuming a two-sided 5 percent significance level, if the mie change in

PIL-CT was 3.080% per unit change in the ADAS-cog score. The value for the true change was

derived from the individual analysis of subject 2 who dernonstrated the most significant

correlation between ADAS-cog score and PIL-CT responses (r=-0.999, p=0.001) and had an

observed rate of change of y=-3.08~+73.85.

Despite the limitations associated with analysis of the result t om this study. this does not

preclude the possibility of a productive discussion of the results. As such, the followinp sections

describe the results of this study and the rnethodology utilized.

In the first part of the study, the pupillary response to pilocarpine of healthy elderly and .AD

patients was compared over a 60-minute time Frame. Our results indicated that healthy rlderly

control subjects had a greater response to PIL-CT than AD patients did (e.g. maximum response

for elderly controls 33.5 f 6.0% vs. 28.6 t 7.3% for AD patients). This is in opposition to

previous studies that were able to detect a significantly greater pupillary response to pilocarpine

in AD patients compared to elderly controls (90-94). For example, Kaneyuki et al. (90)

measured changes in pupil diameter kom baseline in response to 2 drops of pilocarpine

(0.0625%, volume not specified) at 5 tirne points over 45 minutes ro determine whether miotic

response to pilocarpine could be used as a diagnostic test to distinguish AD patients from healthy

elderly. (As a side note, they dso tried to replicate the original tropicarnide study, but were

unable to detect a significant difference between the 2 groups.) Their results revealed that the

AD patient had a significantly greater decrease in pupil diameter compared to the control

subjects (p c 0.001). The authon found a minimum overlap of response at 15 minutes post eye

drops, where 85.7% (12A4) of AD patients had a decrease in pupil diameter 2 23% and 904'0

(Y1 0) control subjects had a response < 23%.

Previously, the question was why individuals with decreased central cholinergic functioning

appear to show a hypersensitive response to pilocarpine and not to tropicamide. Some possible

explanations included: 1 ) differences in absorption due to differences in the molecular sizr luid

charge of the two drugs; 2) differences in mechanisms of action between agonists and

antagonists; and 3) differences in receptor specificity of pilocarpine and tropicamide for various

muscarinic receptor subtypes and changes in various subtype populations in aging. .Alternative

explanations include either methodological considerations or individual variations.

Methodological factors include differences in the eye drops (e.g. prepmtion, volume.

concentration. administration, etc.), pupil measurement (e.g. tec hnology, sarnpling times.

conditions. etc.), and analysis (e.g. placebo-controlled, % change from baseline. etc.). Individual

issues include physiological differences such as variability in comeal permeability, drynrss of

eyes. tear production, blink rate, colour of the iris, and genetic factors. al1 of which could affect

the absorption of the eye drop and thus the observed response.

One genetic factor that appears to be important is apolipoprotein E (ApoE) genotype. ApoE is

an important plasma protein that acts as a modulator of cholesterol and phospholipid transport

between different ce11 types (125). ApoE has 3 alleles: ~ 2 , ~ 3 , and ~ 4 . Though the frequency of

the €4 allele is low arnong the general population' it is markedly higher in patients with AD

( 126). S tudies invesûgating the association of pupillary response to tropicamide and the ApoE

allele showed that healthy individuals with the ApoE €4 allele had a hypenensitive response to

tropicamide (p<0.0001) that was significantly greater than those without the allele ( 177). I t was

also found that AD patients homozygous for the ApoE ~4 allele also demonstrated a greater

response to tropicamide than heterozygotes or those without the allele ( 1 30). The mechanism by

which ApoE genotype appears to influence pupillary response to choiinergic challenge has yet to

be rlucidated. What is clear though, is that the ApoE ~4 allele may be an important determinant

in ocular testing. However, it is not knomn how to account for the effect of this allele. It is

conceivable that the differences in the AD diagnostic studies were driven in part by the presencr

or absence of the ApoE EJ allele in the AD and control populations. There have also not been

any studies that have explored the relationship between ApoE genotype and pupillary response to

pilocarpine. Though it is likely that the ApoE ~1 allele also increases the mioric sffect of

pilocarpine, they may not interact.

Our results show a trend that healthy elderly subjects have a larger pupillary response than .a

subjects; however, due to the limited sample size, this may have arisen purely due to chance.

This trend was also seen in one of the tropicamide studies in which the control group (n=?J) had

the greater response than the AD group (n=20) (128). Interestingly. the Kaneyuki study (90)

resembles the Scinto tropicamide study (23) in that it appears to have an under-responsive

control group. As mentioned in the Background, the subsequent tropicamide studies were not

able to replicate the results of the original study not because the AD patients had a lesser

response than Scinto's, but because of a greater response in the control subjects, similar to XD

patients. In the same way, Kaneyuki's AD patients had a response of -35% compared to -16%

for controls, whereas healthy elderly subjects in 2 other studies had response to pilocarpine

between 30.7% to 34% (25, 129). (In one of the studies (25), control subjects received

pretreatment with tropicamide; however, as tropicamide opposes the miotic effect of pilocarpine.

we would expect that the response to pilocarpine would have been even larger had tropicarnide

not been administered.)

We have never believed that pupillary response to cholinergic challenge could be used as a

diagnostic test for AD. Rather, we proposed only the responsiveness of the periphrral

cholinergic system of the iris rnay be an index of central cholinergic functioning. One rnay argue

that this distinction is merely semantics; however, it is becoming more clear that though central

cholinergic pathology rnay be a major charactenstic of the disease, there is considerable

heterogeneity in AD that includes marked alterations in several other brain neurotransmittcr

systems. There is also the issue of diagnostic threshold such that central cholinergic deficits rnay

be masked by cognitive reserve secondary to education level or innate IQ in some individuals. .A

simple. non-invasive probe of the central cholinergic system would have many potential

applications in the treatment and monitoring of cholinergic diseases such as AD.

In the second part olthis study, the utility of PIL-CT to predict treatrnent response to the ChEI

donepezil in AD was evaluated. Responders measured by the ADAS-cog, MMSE. and DAD had

a trend towards a lower baseline PIL-CT response. This supports the postulate that individuals

with greater central cholinergic deficits rnay be less able to respond to ChEI therapy ( 12, 1 1 1 ). .A

correlation (r=0.72), though not significant @=O. 17) was seen between baseline P IL-CT scores

and improvement in MMSE scores in accordance with this theory. However, the opposite trend

was seen with the CIBIS+ where the association appears to exist between improvement and

~ e a t e r cholinergic decline. The response of subject 5, who had an abnormally low PL-CT

score at baseline, rnay explain these hdings. The ADAS-cog, MMSE, and DAD rated subject 5

as a responder, but the CIBIS+ rated subject 5 as a non-responder. Due to the small sample size.

the mean baseline PIL-CT scores for these groups appears to be driven by the response of subject

5. The correlation between PIL-CT response and MMSE also disappears if subject 5 is removed

from the analysis.

PIL-CT responses were aiso exarnined to see if they were an accurate representation of disease

state or disease progression. PIL-CT scores were plotted versus ADAS-cog scores fiom ail

sessions dunng which both tests were performed (including the results For subject 1. who tiad

completed baseline assessments). As a group. no correlation was Found (r = -0.05, p = 0.85): but.

when PIL-CT responses were plotted against ADAS-cog scores for each subject individually.

trends were observed in 4 of 5 subjects, but the correlation was only significant in I subject

(subject 2; r = -0.999, p = 0.001). In 3 subjects, increases in cognition as rneasured by the

ADAS-cog were accompanied by a reduction in PIL-CT responsiveness (increase in periphsral

cholinergic function). As well, in several instances, a decrease in cognition tiom session to

session was accompanied by an increase in PIL-CT score (decrease in penpheral cholinergic

function). These observations were consistent with the hypothesis that central cholinergic

bnction may be measured by a penpheral probe. This was also seen in the group as a whole as a

4.8 + 1.6-point improvement in ADAS-cog was accompanied by a 13.3 k 7.3% decrease in PIL-

CT response. Subject 5 showed the opposite trend in ihat PIL-CT increased as ADAS-cog

improved, but again, subject 5 had abnomal baseline response, which may have had a profound

influence on the results. Subject 6 had no correlation between responses; however, a ceiling

effect may have occurred with the ADAS-cog as subject 6 scored very highly at baseline, leaving

little room for measurable improvements.

This study was another attempt at using pupillary response to the penpheral cholinergic system

of the iris to distinguish individuals with different degrees of centrai cholinergic function. This

was also the second study that examined the correlation between these responses. PIL-CT has

had mixed results as diagnostic test of AD or a predictor of treatment response to C E 1 therapy.

Pupillary responses have also not been well correlated with measures of the central cholinergic

sysrem. For this stuciy, again, the srnail sampie size, which fell weil beiow the projected nurnber.

must be taken into account when interpreting the results. However, a significant correlation aas

present in one individual in this study and when so called "idiosyncratic" responsrs were

removed From the analysis of another study (25). These pockets of evidence support the

hypothesis that there is likely some relationship between the central cholinergic system and the

pupillary peripheral cholinergic system. Unfortunately, although it appears that such an

association may exist, the methodology may be lacking to measure and apply this in any clinical

uselul way. Several challenges have been observed that seem to involve the execution of the

methodology, which has been problematic, rather than flaws in the theory behind the study itself.

These issues are outlined below.

1 ) Eye Drops

Although substantial attention was given to the method of adrninistering the eyedrops in this

study, there are several problems that can occur during the procedure. Many considerations were

implemented to "minimize" these complications (Methods, page 19); however, complications

may have occurred which would have had a profond effect of the observed responses. The

most important and potentially serious problem was the "blink reflex". During a nurnber of

sessions, the subject blinked as an eye drop was instilled. This may have an effect on the amount

of cimg reaching the site of absorption and measured dmg effect, but the extent of this

occurrence is not quantifiable.

When a "blink reflex" is observed, very little can be done. A second eye drop cannot be

adrninistered, yet it is probable that the fint drop did not reach its target completely and the

outcome of the session will not be an accurate result. Idedly, the data from the visit should be

discarded and another session run in its place, but this was not Feasible. The consequences of the

"blink reflex" c m be illustrated by examinine the results of subject 5. This subject was generally

quite anvious during the instillation of the eye drops and as a result would blink O Ften while

awaiting the drops. It was specifically recorded that the subject blinked during the delivery of

the eye drops during the baseline session. The analysis of this session revealed that the

maximum decrease in pupil diameter was 6.3%. This response was less than what would be

expected for a healthy young subject (25). in subject 5's second session. there was a rnarked

improvernent in ADAS-cog score, but a substantial increase in responsiveness to pilocarpine. I t

is possible that these changes may not be related and the hypothesis that there is a link between

central and peripheral cholinergic response is incorrect; however. it seems more likely that the

results from subject 5's baseline session was not accurate due to the "blink reflex". If this value

was not correct, it would affect the analyses of individual changes and correlations. as well as the

group analyses.

One possible solution that has been used is to spread the delivery of the drug to 2 eye drops

instead of the single administration (90. 93, 130). This is surely only a partial answer. though.

Using 2 drops may provide an extra opportunity to administer the dmg to its target receptors. but

it also presents an additional o p p o d t y for Iess than optimal instillation. If a blink did occur

during application of one of the 2 drops (or both!), it may reduce loss of drug, but would not

eliminate it completely.

1) intra-individual and inter-individual variation

Lnm-individual vanability is also a potential limitation to pupillometry. One of the tropicamide

studies examined test-retest reliability and the results should be a cause for concem. Csing a

classification of a 13% increase in pupil diarneter fkom baseline at 30 minutes to classi fj siibjects

as "positive" or "negative" responden to tropicamide, the authors found that over 50% of

subjects changed groups on a second test (30156) compared to their first test ( 13 1 ).

The degree of dmg reaching the site of action is dependent on a plethora of physiological

characteristics including comeal permeability, dryness of eyes, tear production. blink rate. and

iris colour (24). These factors seldom remain constant f?om moment to moment in an individual

and will al1 have an effect on drug absorption. Thus, assuming complete delivery of an eye drop

to the surface of' the cornea, the measured response may Vary considerably depending on timing

of the test. Steps were taken to reduce variation, such as testing at the same time of day. in the

sarne room, etc, but many conditions could not be controlled. Things like weather. arousal.

emotional state, general health, dust in eye, etc. and their effect on the ocular state changed

throughout the day and also from session to session. This made it exceedingly difficult to be

sure that absolute pupillary measurements and change scores are accurate.

h tuni, inter-individual variations in subjects will no doubt have an effect on the rate and extent

of h g absorption and subsequent response. This would raise questions conceming the validity

of any analyses involving comparing subjects as it woula be difficult to determine if the

responses being compared resulted from equal amounts of dmg. This study attempted to account

for variability in rate of absorption by using maximum response for analysis rather than

responses taken at discreet time points. However, there were limits to controlling the extent of

absorption. The volume of a typical eye drop is 50 pL (13 l ) , but the volume of the lacrimal lake

in the cul-de-sac of an upnght, blinking patient is only 7 to 1 O pL (132, 133). A smaller volume

drop was used (20 PL) and subjects were placed in the supine position dunng application of the

rye drops, but individual variations in the volume of the pouch formed by the lower eyelid would

influence the amount of dmg retained. Subjects were also instructed to apply disita1 pressure io

the nasolacrimal duct, but inevitably drainage may occur and the degree of elimination may Vary.

The duration that dnig is in contact with the comea is critical in determining the extent of

absorption that occurs. This length of time is affected by aforementioned charactsristics. For

example, one study exarnined the rate of blinking between Alzheimer's disease patients and

controls and Found that aithough there was a difference between the two groups (18 + 17

blinkslmin vs. 13 ?: 10 biinks/min), it was not statistically significant (p=0.32) (134). Despitr

this finding, it is clear that there are differences in this factor between individuals (oRen large

di fferences), which may affect pupillary response.

3) The Pupillometer Apparatus

The original unilateral pupillometer was sirnply a video carnera hooked up to a video monitor

and a personal computer with software that was developed to calculate pupil diameter. During

the baseline visit with subject 1, a hardware breakdown caused the session to be voided. The

pupiilorneter was subsequently repaired and the session was completed. Mid-way through the

study, the pupillometer ceased to function again. Despite consultation with the institution's

technicians, the apparatus was not repaired and session 3 for subject 4 was missed. Fortunately.

another pupillometer was found, but it was not without its share of problems. The second

pupillometer was bilateral in design (2 video cameras mounted in parallel), but it proved quite

challenging to rapidly focus the images of both eyes simultaneously. The pupillometer was not

onginally designed for long terni use and was particularly fiagile and required delicate handling.

As the length of time required to capture the pupillary image increased, the attention of the

subject (particularly the AD patients) would begin to wander and the subjects would often take

their focus off the standardized target. Pupillary accommodation was observed during testins as

the subject presumably began "looking around". Subjects were always re-instructed to focus on

the target again. before the image was taken. but it was difticult to determine whether the

changes in pupil size due to the visual accommodation had any affect on the results. Thus. use of

both pupillometers proved to be problematic.

4) A Lack of Central Cholinergie Measures

The first 3 sections describe difficulties with assessing penpheral cholinergic function; however.

another limitation to these studies is the lack of a method of accurately measuring central

cholinergic function. Previously, scopolamine has been used as a probe of the central

cholinergic system, but it can be undesirabie because of the central cholinergic side effects

associated with its use. Particularly in elderly and AD patients, scopolamine is known to cause

considerable confusion and disorientation. In the instances where scopolamine has been used. a

measure of cognition still needs to be used to detect changes in central cholinergic function. In

the past the Buschke selective reminding test (SRT), or some variation, has been used as a

cognitive outcome. However, our study was not able to correlate central cholinergic responses

as measured by the SRT and penpheral cholinergic response to pilocarpine (25).

in this study, the standard eficacy assessments (ADAS-cog, MMSE, DAD, and CIBIS-) used

by regulatory agencies for ChEIs were use to evaluate central cholinergic function and response

to donepezil. These scales are designed to obtain a global rating of cognition and function.

thouph, and are not focused solely on cholinergic components. Currently, there are no validated

scales of central cholinergic activity, thus these and other scales will continue to be employed in

the interim. ln actuality, central and penpheral cholinergic responses may not be associated. but

then it rnay be the methods used to measure central cholinergic function that limit our ability to

detect a correlation.

5) A Lack of Clinical Utility

Despite evidence ihat the peripheral cholinergic system of the iris rnay reflect central cholinersic

function, it is unlikely that this procedure will achieve any significant clinical utility. The

inability of pupillary challenge to serve as a diagnostic test for AD has been clearly

demonstrated. This may be due to the limitations in accurately measuring penpheral choiinersic

responses, the complex nature of the disease itself, or the possibility that changes in central and

penpheral cholinergic responses seen in cholinergic decline are not closely linked.

Currently, al1 suitable candidates for ChEI therapy are placed on donepezil therapy as early alter

diagnosis as possible, as there are no other treatment options. Unless a test could classify

patients as responders and non-responden with near perfect accuracy, most individuals would

likely opt for a ChEI trial anyway. There is also no reason why as other non-cholinergie

treatments anse that these novel therapies could not be used concomitantly with a ChEI. The

question of how to classify patients anses, as well. Wiîh al1 the inherent limitations to

comparing PL-CT responses, even if cutoff scores existed, the procedure has not demonstrated

the precision and accuracy required for this method to be accepted as reliable and valid.

PIL-CT shows the greatest promise as tool to measure disease state or disease progression.

Changes in central cholinergic function due to ChEI treatment, as rneasured by cognitive testins,

appear to be accompanied by changes in penpheral cholinergic system. However. this was only

seen in certain individuals and not when exarnining the treatment group as a whole. Thus. P L -

CT response may reflect the disease state on an individual level. Unfominately. if this is the

case. cognitive testing would seem to be a much simpler method of monitoring AD progression.

Despite the small sample size and the limitations associate with this study, the results provide

funher evidence that changes in the central cholinergic system are accompanied by changes in

the peripheral cholinergic system of the iris. However, due to the lack of statisticai power. any

trends observed must be interpreted with caution. Clearly, there is a need for lunhrr research in

this area to clarifi the exact nature of the relationship between cholinergic responses and develop

clinically useful applications for this procedure; however there are limitations associatrd with

ocular cholinergic challenge and pupillometry which need to be addressed before this research

continues.

The number of subjects participating in this study was well below the targeted sample size. This

limited the power of the study and the statistical analyses that accompanied the results. The

foilowing are putative results based on the descriptive trends observed in the study.

1. Healthy elderly patients had a greater pupillary response to pilocarpine eye drops than .AD

patients did (33.5 f 6.0% vs. 28.6 + 7.3%. respectively).

2. Increased central cholinergic function during ChEI phmacotherapy with donepezil.

demonstrated by improvements in cognitive performance (4.8 ? 1.6-point improvement in

ADAS-cog), is accornpanied by decreased penpheral cholinergic responsiveness of the iris to

pilocarpine (1 3.3 k 7.3% decrease in PIL-CT).

3. Treatment responses to donepezil did not conelated with baseline, pre-ueatment response ro

ocular pilocarpine challenge.

4. Changes in central cholinergic activity as measured by the ADAS-cog did correlate with

changes in peripheral cholinergic responses of the iris in certain individuals.

5 . Evidence continues to accumulate to support the link behveen central and prriphcnl

c ho linegic responses, however, there are several limitations to the tec hno logy present l y

employed to study them which may require further development.

STUDY 2 - EPFICACY AND SAFETY OF ORAL CHOLINESTERASE INHIBITORS IN ALZHEIMER'S DISEASE

There have been a number of pivotal randomized controlled trials demonstrating the efficacy and

tolerability of oral ChEIs ( 19. 20, 105, 106, 114, 135- 139). However, the extent of the benefit

offered by ChEIs, though statistically significant, has been described as moderate at best ( 13.

140-142). A search of the literature revealed only 2 attempts to combine the results of these

trials using meta-analysis, a statistical technique often employed to quantitatively incorponte

outcomes of different studies (143). The first article pooled the results of randomized controllrd

tacrine trials and concluded that its use was limited by withdrawals due to elevated liver enzymes

and that "the clinical relevance of the benefits of (tacrine) remains controversial" ( 17). The

second study determined the numbers needed to treat with appropriate doses of ChEIs to

demonstrate a clinical improvement or a slowing in cognitive decline (144). The results

indicated that srna11 nurnbers of patients (between 3 to 7 for most cases) needed to be treated.

However, the inclusion criteria were set such that only 5 studies were included in the analyses.

As the prevalence of AD continues to nse, a comprehensive meta-analysis of the ChEI literature

would serve to clariw the benefits associated with ChEI phannacotherapy. The tolerability of

ChEI is also an important consideration, as seen in the tacrine meta-analysis, and should

accompany any study of ChEl efficacy. Furthemore, a more accurate accounting of the

proportion of responders and non-responders to ChEIs may support the need for predicton of

treatment response.

METHODS

Literature Search

A search of the English literature kom January 1980 to June ZOO0 was conducted to identify

randomized controlled trials of oral cholinesterase inhibitors in the treatment of Alzheimer's

disease. Potential papers were found by searching Medline (key words: cho linesterase inhibi tor

AND Alzheimer; limits: randomized controlled trials, English, hurnan) and the Cochrane

database. Searches were also conducted for individual cholinesterase inhibitors (kry words:

donepezil. E2020, or Ancept; tacrine, THA, or Cognex; velnacrine, HP 029. or !Mentant.:

rivastipine. ENA 713, or Exelon; metrifonate; galantamine or galanthamine; physostigmine:

eptastigmine; .AND Alzheimer). Recent review articles and potential papers were manually

cross-referenced and experts were consulted for additional articles.

Inclusion/Exclusion Criteria

The following cntena were necessary to include a paper in the meta-analysis:

original clinical trial of a cholinesterase inhibitor in AD

randomized

4 double-blind

placebo-controlled

parallel group

oral mode of administration of treatrnent

0 cognitive outcornes measured by any sa l e

length of treatment at least 12 weeks

Papen were excluded if the raters were not able to extract data for at least one of the four fields

of required information. The required information consisted of the following:

the proportion of subjects who responded to each treatment

the proportion of subjects who expenenced any adverse event

the proportion subjects who dropped out of the trial For any reason

the proportion of subjects who dropped out of the trial due to an adverse event

Assessrnent of Studies

Due to the large number of potential papers, a preliminary screening for inclusion critena w s

perforrned to reduce the number papers required for expert review. The remaining articles were:

1) evaluated for inclusion; 2) quality assessment; and 3) data extraction. "Differentiai"

photocopying was used to blind raters to the authors, locations, dates. and journal of publication

throughout the above-mentioned 3 phases to reduce potential sources of bias. A Medline search

was perfonned with no language restrictions to determine if any additional articles could be

found. Three additional papers were identified, but all three failed to meet inclusion citeria

(145-147).

Assessrnent of Inclusion

The Methods sections of the 137 papers found were reviewed by 2 raters (K.K.Y .. PM. ) for the

criteria necessary for inclusion. Of these studies, 30 were m e r examined by 3 expert raters

(N.H., K.L.L., B.A.L.): each of these raters received 6 to 8 papen to review individually, as well

as 10 papen chosen at randorn to determine inter-rater reliability (Appendix C).

Qdiy rissessrnent

Each of the expert raters received the "differentially" photocopied Methods and Results sections

of the 25 articles that met the inclusion criteria. The description of the snidy subjects. trial

design, results, and data analysis were rated to assess the quality of each study. Each attribute

could be assigned a score of 3 (good), 2 (satisfactory), 1 (poor), or O (not reported). An overall

quality rating, independent of previous scores, was also given tor each paper (Appendix D).

Two analyses were performed to determine the effect of quality on outcorne. In the tirst

analysis, poor papers were removed (i.e. median overall quality 51) and outcomes were

recalculated. In the second analysis, each paper was weighted by quality and outcomes were

recalculated. Median quality score was used in these analyses because the numeric average of

the categorical scale had no meaning.

Dam E-riraction

Each of the expert raters received the Results section of the included studies and extracted the

lollowing data: (Appendix E)

the number of respondershon-responders to treatment

cognitive responders were defined as subjects with a &point improvernent in the

Alzheimer's Disease Assessrnent Scale, cognitive portion (ADAS-cog)

global assessment responden were defïned as subjects rated as improved by a global

assessment scale (e.g. Clinician's Interview Based Impression of Disease Severity with

Caregiver Input (CIBIS+))

a the number of subjects reporting any adverse event

O the number of dropouts (due to any cause)

the number of dropouts due to adverse events

Data Analvsis

The effect sizes of the 25 papers included in the rneta-analysis were cornbined according to the

method of Dersirnonian and Laird (148) with the modification by Velanovich (149). This meta-

analytical statistical technique calculated outcornes by weighmg each study by the number of

subjects and variance. Pooled mean percentages of "responders" (to treatment, adverse events.

etc.) and 32% confidence intervais [95% Cij were cietermined. For exarnpie. for a CiiEi and

placebo:

I I ChEI

The rates of response for the ChEI and placebo groups and the difference berween groups For

each study were calculated:

Placebo 1 C

Calculate the rate of response for the ChEI for each comparative study (i)

R A = A / ( A + B )

Calculate the rate of response for placebo for each comparative study ( i )

R e = C / ( C + D )

Calculate the rate of difference between the ChEI and placebo for each comparative study

RDIFF = RA - RB

Responders

A

D !

The variances for each of the ChEl and placebo groups and the combined variance for each study

were calculated:

Non-responders

B 1

1. Calculate the variance for the ChEI for each comparative study (var) (variance within each smdy)

5. Calculate the variance for placebo for each comparative study (var) (variance within each study)

6. Calculate the combined variance for both the ChEI and placebo for each comparative study (varcomb) (variance within each study)

The combined variances were then used to calculate the weight and weighted effect size for each

of the studies, and the weighted sumrnary of effect sizes for al1 the studies:

7. Calculate the study weight for each study (i) (w,)

w, = 1 / vaRancecornb = 1 / V m o m b

8. Calculate the weighted effect size for the rate difference for each comparative study ( i ) (wR,)

WRDIFF = RDIFF x wi

9. Calculate the weighted summary of the effect size for al1 studies (R*)

R* = Sum of study weighted difference effect sizes 1 Sum of study weights

R* = wRDIFF 1 Wi

The homogeneity of the studies was calculated:

10. Calculate Homogeneity Q (X'-statistic)

Q = E [wi x (RD~FF - R*)']

The variance of the weighted summary of effect sizes for d l studies (R*) was calculated:

1 1. Calculate the variance of R* (between studies) (var R*)

varR*=([Q-@-l)]/[Xwi-(~wiZ/~wi)]} (ifvarR*<O,varR*=O)

The variance within and between studies (Le. the "double" weight was calculated to detemine

the weight of the difference in rate of response for each study and the final effect size:

12. Calculate the "double" weight (wi*) (weight of the variance within and between studies)

y* = 1 1 (wi + var R*)

13. Calculate the weight of each Ri (wi *Ri)

W*RDIFF = wi* x RD~FF

14. Calculate the final effect size (Effect size = Rate of Response)

E ffect size = x ~ i * R o i ~ ~ i I w i *

The standard error and confidence intervals for the final effect size were calcdated:

15. Calculate standard error of the final effect size (SE)

SE = 1 1 [\l(Xwi*)]

16. Calculate the 95% Confidence Interval (95% CI)

CI = Effect size k (1.96 x SE)

Pooled mean percentages were also calculated for each of the data extracted in a single arm

analyses for al1 ChEIs, placebo, and dosage of ChEIs. Dosages were assigned to the Iow. mid.

and high dose groups according to common prescribing practice ancilor analysis of the literature

(Table 3). An objective mesure such as peripheral AChE inhibition could noi be used to

compare doses between drugs due to differences in potency of central and peripheral

cholinesterase inhibition (e.g. galantamine is -10 fold less potent at inhibiting brain AChE than

red blood ce11 AChE) (141). Studies were a h grouped by first generation ChEIs, characterized

by short duration of action and lack of specificity for AChE (tacrine, velnacnne, and

physostigmine) and second generation ChEIs, longer duration of action and increased specificity

(donepezil. rivastigmine, eptastigmine, galantamine, and metrifonate) ( 142). The heterogeneity

behveen the studies combined was calculated using the method of Breslow and Day. which

determines a Q value that follows a X' distribution (1 50).

Table 3 - Summary of ChEl dosage classification

ChEI Low Dose Mid Dose High Dose

Donepezil

Eptastigrnine

Galantamine

Metri fonate

P hysostigmine

Rivastigmine

Tacrine

Velnacrine

S tudies

2 raters initially excluded 107 studies (58. 106, 121, 123, 124, 142. 151-252) for the following

reasons: not a ck!ica! tria! (n=30), not an origina! c!hical Vk! (n=?!), not randomized (n-! ). no[

controlled (n=2). not double blind (n=18), not parallel (Le. crossover) (n=29), ChEL treatment not

oral (n=10). insufficient length of treatment ( c l 2 weeks) (n=j 1). and/or no cognitive outcornes

(n=2 1 ).

The remaining 30 studies reviewed by the expert raten, with 25 studies meeting inclusion

criteria (1 5. 19-22, 105, 135-139. 253-266) and al1 but one (253) had extractable data (Table 4).

These studies examined ChEI therapy in 7061 subjects and placebo treatment in 3256 subjects.

Of the 5 studies excluded by the 3 expert raten (267-271), the reasons for exclusion included:

not original clinical trial (n=3), not double-blind (n4). and not controlled (n=l ). The extractrd

data from the 24 studies included in the meta-analysis are summary in Table 5 .

Comparative analyses

Eflcacy Analyses

The proportion of subjects demonstrating a significant cognition improvernent (+point

improvernent on ADAS-cog) could be extracted from 7 of 24 studies involving 3307 subjects.

The pooled mean percentage of subjects who showed a significant improvernent in ADAS-cog

score was 31% [95% CI: 24% to 38%] for ChEIs and 18% [95% CI: 16% to 21%] for placebo.

-

Table 4 - Studies meeting inclusion critena for meta-analysis of oral cholinesterase inhibitors in Alzheimer's disease

Au thor(s), Y ear C holinesterase Dose Number of Patients Outcorne Duration of Inhibitor (ChEI, placebo) Measures Treatment

- - - . . .

Farlow et al., 1992 tacn rie 20-40mg 468 (r) 273 (c)

(39 1,77)

Wood & Castiederi, 1994 tacrine 20- 120nig 154 (r) 123 (c)

(78,76)

Knapp et al., 1994

Autuono, 1995

tacrine 80- 1 60mg 663 (r) 379 (c)

(479, 184)

velnacrine 150-225mg 449 (r) 280 ( c )

(297, 152)

Becker et al., 1996 nietri fonate 2. l mg/kg 51 (0 50 (cl

(27,23)

Rogers et al ., 1 996 doiiepezi 1 1 -5~iig 161 (r) 141 (c)

(121,40)

Inibinlbo et al., 1998 eptiisiigiiiine 30-45 mg 320 (r) 286 ( c )

(2 14, 1 OU)

ADAS-cog 12 weeks CGIC

MMSE 12 weeks CGRS

ADAS-cog 30 weeks CIBIS+

ADAS-cog 24 weeks CGIC'

ADAS-cog 3 montlis

ADAS-cog 12 weeks CGIC

ADAS-tog 25 weeks CIBIS t

Table 4 - Studies meeting inclusion criteria for meta-analysis of oral chohesterase inhibitors in Alzheimcfs disease (cont) - -- -. . .

Author@), Year . . . . - -- - - -

Cholinesterase Dose Number of Patients Ou tco~ne Inhibitor (ChEI, placebo) Measures

Duratioii of Treatrnent

Cumrnings et al., 1998

Corey-Bloom et al., 1998

Agid et al., 1998

Becker et al., 1998

Rogers et al., 1998b

Rogers et al., 1 W8a

Morris et al,, 1998

- - -

metri fonate

rivastigmine

rivastigmine

metn fonate

donepezil

doiiepezi l

inetri fonate

- .-

1 O-6Omg 480 (r) 433 (c)

(360, 120)

1-12mg 699 (r) 544 (c)

(464,235)

4-6nig 403 (r) 357 ( c )

(269, 133)

5- 1 Omg 468 (r) 412 (c)

(315, 153)

5- 1 Onig 473 (r) 368 (c)

(31 1 , 162)

30-O(hg 408 (r) 334 (c)

(273, 135) - - - - - - - - - - - - - - - - --

ADAS-cog MMSE

. . . .. . -.

12 weeks

26 weeks

13 weeks

6 inonths

12 weeks

24 weeks

20 weeks

Table 4 - Studies meeting inclusion criteria for rneta-analysis of oral cholinesterase inhibitors in Alzheimar's disease (cont)

Au thor(s), Y ear C holinesterase Dose Number of Patients Outcoine Duration of Inhibitor (ChEI, placebo) Measures Treatment

Imbimbo et al., 1999 epiastigmine

Rosler et al., 1999 rivastigmine

Thal et al., 1999 pliysost igmine

Wong et al., 1 999 tacrine

Burns et al., 1999 donepezil

Dubois et al., 1 999 nietri fonate

Raskind et al., 1999 iiictri fonate

- . . -- - - - -- - - - - - -- . - . - - - .

491 (r) 424 (c)

(327, 164)

725 (r) 581 (c)

(486,239)

475 (r) 210 (c)

(358, 117)

100 (r) 56 (cl

(75,25)

818 (r) 631 (c)

(544,274)

605 (r) 5 10 (c)

(397,208)

260 (r ) 2 10 (c)

( i 79, S7) - -. . . . - . - -. . . -. . . - . -. . . . - . .- - -

CASI MMSE

24 weeks

26 weeks

24 weeks

30 weeks

24 weeks

24 weeks

26 weeks

- Table 4 - Studies meeting inclusion critena for meta-analysis of oral cholinesterase inhibitors in Alzheimer's disease (cont)

Author(s), Year Cholines terase Dose Number of Patients Outcome Duration of Inhibitor (ChEI, placebo) Measu res Treatment

Van Dyck et al., 2000 physostigmine 24-30mg 1 76 (r) ADAS-cog 12 weeks 148 ( c ) CIBIS+

(83,93)

Raskind et al., 2000 galantamine 24-32rng 636 (r) ADAS-cog 6 nionths 438 (c) CIBIS+

(423,2 13)

Tariot et al., 2000 galantaniine 8-24 mg 978 (r) ADAS-cog 5 nionths 779 (c) CIBIS+

(692,286)

Abbreviations for Table 1 : (r) = number of subjects randomized; (c) = coiiipletors (Le. nuniber of subjects wlio conipleted the study); ADAS-cog = Alzheimer's Disease Assessment Scale, cognition portion; CIBIS+ = Cliniciaii's Interview Ilased Impression of Disease State with cnregiver input; CGIC = Clinician's Global Impression of Change.

- - - - . . . . .

Table 5 - Data extracted fiom studies included in meta-analysis of oral cholinesterase inhibitors in Alzheimer's disease -- -

Au thor(s) Cholinesterase Proportion Proportion Proportion Proportion of Proportion of Median Year Inhibitor lmproved lmproved with Adverse Dropouts Dropouts due to Quality

ADAS-cog CIBIS+ Even ts Adverse Events Score (1 to 3)

Farlow et al., 1992

Wood & Castleden, 1994

Knapp et al., 1994

Autuono, 1995

Becker et al., 1996

Rogers et al., 1996

Imbimbo et al., 1998

Cummings et al., 1998

Corey-Blooni el al,, 1993

Tacri ne Placebo

Tacri ne Placebo

Tacrine Placebo

Velnacrine Placebo

Metri fonate Placebo

Donepezi 1 Placebo

Eptastigmine Placebo

Metri fonate Placebo

Rivasi igiiiine Placcbo - - . . . - - . .

Table 5 - Data extracted from studies included in meta-analysis of oral cholinesterase inhibitors in Alzheimer's disease (cont)

Au thor(s) Cholinesterase Proportion Proportion Proportion Proportion of Proportion of Median Y ear Inhibitor Improved lmproved with Adverse Dropouts Dropouts due to Quality

ADAS-cog CIBIS+ Even ts Adverse Events Score (1 to 3)

Agid et al., 1998

Becker et al,, 1998

Rogers et al., 1 W8a

Rogers et al., 1998b

Moms et al., 1998

lmbimbo et al., 1999

R6sler et al., 1999

Wong et al., 1999

Rivasiigmine Placebo

Metri fonate Placebo

Donepezil Placebo

Donepezil Placebo

Metri fonate Placebo

Epiast igniine Placebo

Rivastigniiiie Placebo

Pliysostigniine Placebo

Tcicriiie Plciccbo

n/a

nla

137/302 41/153

nla

d a

ida

931484 39/23 8

801303 1 2/80

l i h

- . .

2001269 3811 33

1 4124 8/23

n/a

23013 15 1061153

1 73 9011 34

1 66/32 7 811164

3031484 1731339

nla

ii/a

-

Table 5 - Data extracted fiom studies included in meta-analysis of oral cholinesterase inhibitors in Alzheimer's disease (cont)

Author(s) Cholinesterase Proportion Proportion Proportion Proportion of Proportion of Median Year Inhibitor Improved Improved with Adverse Dropouts Dropouts due to Quality

ADAS-cog CIBIS+ Events Adverse Events Score (1 to 3)

Burns et al., 1999

Dubois et al., 1999

Raskind et al., 1999

Van Dyck et al., 2000

Raskind et al,, 2000

Tariot et al., 2000

Donepezil Placebo

Metn fonate Placebo

Metri fonate Placebo

Physostigmine Placebo

Galantami ne Placebo

Gülantaniine Placebo

d a

nla

nla

d a

83/248 261 157

152/419 44/225

125/544 3 8/274

n/a

nia

n/a

64/3 99 271207

nia

Abbreviations: ADAS-cog = Alzheimer's Disease Assessiiient Scale, cognition port ion; CIBIS+ = Cliiiiciiin's Interview Based Impression of Discase State wiili caregiver input; nla = iiot avüilable/unable io exirüct.

The therapeutic eMect (ChEI minus placebo) was 12% (95% CI: 7% to 17%) (Table 6) (Figure

11). The proportion of subjects showing an improvement in global assessment of change

(improvement on CLBIS+/CGIC) could be extracted from 10 of 24 studies involving 4905 of

subjects. The pooled mean percentage of subjects who showed a significant improvement in a

global assessment scale was 27% [95% CI: 23% to 31%] for ChEIs and 17% [95% CI: 1-196- to

i 9?o] for placebo. The herapeuiic eifect was 9% [95% Ci: 7% to 1Z%j (Figure II). Quality-

weighted mean differences were also detemined and were similar in magnitude to rau-

differences. Heterogeneity was statistically significant for the cognitive (Q= 16.1. p=0.0 1 )

analysis. For the cognitive analysis. the heterogeneity was due to one study by Rosler et al.

(135) examining high and low dose rivastigmine. The proportion of patients who had a

significant response to treatment was low ( 19%). Removal of that study from the analysis r a i d

the therapeutic response to 14% [9S0h CI: 1 1% to 18%] and eliminated heterogeneity (Q=5.0.

p=O.41). For the global assessment analysis, the studies were hornogeneous (Q=10.75. p=0.29)

(Table 3). (Please see Appendix F for further details concerning these analyses.)

Safety ha&sis

Rates for advene events, dropouts, and dropouts due to adverse events were also compared

(Table 6). The proportion of subjects experiencing any treatment ernergent adverse event could

be extracted from 15 of 24 studies. The analyses revealed that ChEI treatment resulted in a

modest but significantly greater fiequency of adverse events, dropouts, and dropouts due to

adverse events than placebo. The pooled rnean percentage of subjects who expenenced an

adverse event 69% [95% CI: 60% to 77%] for ChEIs and 61% [95% CI: 48% to 7461 for

placebo. The therapeutic difference was 7% [95% CI: 2% to 12%]. For dropouts and dropouts

due to adverse events, the pooled mean differences were 10% [95% CI: 7% to 14%] and 1 19.0

Table 6 - Summary of Comparative Outcornes for AI1 Extracted Data including Quality Analyses jiom meta-analysis of ChEI treatment of Alzheimer's disease

-- -

Di fference in Di fference in Quality- % lmproved % lmproved Weighted

Total Subjects f Standard w/o poor Mean (ChEl, CliEI Placebo Error 2-score Heterogeneity quality studies Difference

Outcome placebo) [95% CI] [95% CI] [95% CI] (p value) Q (p value) [95% CI] [95% CI]

0.3 1 O. 18 [0.24,0.38] [O. IO, 0.2 1 ]

Adverse Events

Dropouts

Dropouts due to Adverse Events

0.27 0.17 [0.23,0.3 1 ] [O. 14,O. 191

0.27 O , 15 [0.20, 0.331 [O. 12, O. 181

+Al1 ChEls + Placebo

1 ChEl minus Placebo 1

O 1 O 20 30 40 50

% lmproved (95% CI)

Figure 11 - Therapeutic Effect of ChEls (Cpoint lmprovement on ADAS-cog) Pooled estirnate with 95% confidence intervals.

+ Al1 ChEls + Placebo

a ChEl minus Placebo

O 1 O 20 30 40 50

% lmproved (95% CI)

Figure 12 - Therapeutic Effect of ChEIs (Improvement on CIBIS+ICGIC) Pooled estimate with 95% confidence intemals.

[95% CI: 6% to 16%]. There was significant heterogeneity between the studies for safety

(Q=60.3, p<0.001), dropouts (Q=125.9, pc0.001), and dropouts due to adverse events (Q=367.0.

p<O.OOl). For the safety analysis, heterogeneity was controlled by studies by van Dyck et al.

(263) and Wood and Castledon (264), which had greater (100%) and lesser (29%) rates of

adverse events in their placebo groups, respectively. With the removai of these outliers. the

remaining studies are homogenous (Q=18.8, p=0.09).

Single Arm Analysis

.-lnri(vsis by Dose

Pooled mean percentages of patients with a 3-point improvement in ADAS-cop score with

placebo, low. mid. high dosages of ChEIs were calculated (Table 7 to 12). Low dose ChEIs

showed a similar efficacy as placebo (21% vs. 18%). The largest rate of response was seen in

the mid dose ChEI group (35%) rather than the high dose ChEI group (31%) (Table 7). For

improvements in global assessment, the meta-analysis revealed comparable resuits for ail 3

dosages (25% to 28%), which were modestly better than placebo (1 7%) (Table 8). The rates of

adverse events between placebo, low, and mid dose ChEIs were similar (62% to 65%) and less

than high dose ChEIs (78%) (Table 9). For placebo and iow dose ChEIs, rates of dropouts and

dropouts due to adverse events were comparable with ( 15% and 6%, respectively), and were

successively higher in the rnid (24% and 15%) and high dose groups (37//0 and 23%) (Table Ir)

and 11). A dose analysis was also performed to determine the rate of dropouts due to adverse

events with only the studies included in the cognitive and global assessment analyses and a

sh i l a r trend to the analysis including al1 the studies was seen (Table 12). For example, for the

studies bom the cognitive analysis the rates were 9% and 7% for placebo low dose ChEI.

Table 7 - Summary of Cognitive Efficacy Data by Dose

Group Number of Studies Ni~iiiber of Patients Eîlicücy Rate [95% CI] Heterogeneity Q (p value)

All ChEIs 7 3356 0.3 2 r0.29, 0.361 60.7 (p < 0.001)

Placebo 7 Il40 0.18 [O. 15,0.21] 8.9 (p = 0.18) n.s.

By Dose

Low Dose 2 422 0.2 1 [0.09, 0.321 8.6 (p = 0.003)

Mid Dose 5 885 0.35 10.3 1.0.401 7.5 (p = O . 1 1) n.s.

Higli Dose 6 1 049 0.3 1 [0.23,0.39] 44.5 (p < 0.00 1 )

By Class

I " Generation 3 758 0.28 [0.25, 0.3 1 ] 1.4 (p = 0.5 1 ) n.s.

2"d Generation 5 1598 0.32 [0.22, 0.421 75.4 (p < 0.00 1)

Table 8 - Summaty of Global Assessrnent Efficacy Data by Dose . --. -

Group Nun~ber of Studies Number of Patients E Fficacy Rate [95% CI] Hcterogaieity Q (p value)

All ChEls 10

Placebo 1 O

By Dose

Law Dose 2

Mid Dose 6

High Dose 7

By Class

1 '' Generation 3

2"d Generatio~i 7

0.28 [0.24,0.32] 63.3 (p < 0.001)

O. 17 [O. 14,0.20] 22.3 (p = 0.008)

0.28 [0.23, 0.321 1 .O (p = 0.3 1) n.s.

0.25 [O. 19, 0.3 1 ] 3 1.6 (p < 0.00 1 )

0.27 [0.20,0.34] 45.4 (p < 0,001)

- -

Table 9 - Summary of Safety Data by Dose

Group Number of Studies Nurnber of Patients Rate of Adverse Events [95% Cl] I-lelerogeneity Q (p value)

All ChEls 15 4590 0.69 [0.60,0.77] 743.6 (p < 0.001)

Placebo 15 2239 0.62 [0.48,0.74] 1041.4 (p < 0.001)

By Dose

Low Dose 7 665 0.62 [0,45,0,80] 114.4 (p<O.OOI)

Mid Dose 9 2 165 0.65 [OSO, 0.811 486.2 (p < 0.00 1 )

High Dose 6 1241 0.78 [0.68,0.88] 139.5 (p < 0.001 )

By Class

1 " Generiition 4 1 027 O. 73 [0.55,0.9 1 ] 256.0 (p c 0.001 )

2""eneration I l 3563 0.67 [0.58, 0.76) 423.9 (p < 0.00 1 ) -- --

--

Table 12 - Summary of Dropout due to Adverse Event Data by Dose

Group Number of Studies Number of Patients Rate of AE Dropouts 195% CI] Heterogeneity Q (p value)

Cognitive Studies

Placebo

Low Dose

Mid Dose

High Dose

0.09 [O.OC, O. i 1 ] 14.1 (p = 0.01)

0.07 [0.04,0.11] 0.8 (p = 0.67) n.s.

0.24 [O. 1 1, 0.371 181.7 (p < 0.001)

0.3 1 [O. 16, 0.461 144.9 (p < 0.001 )

Global Assessrnent Studies

Placebo 7 141 1 0.06 [0.04,0.08] 20. I (p = 0.002)

Low Dose

Mid Dose

High Dose 6 1277 0.20 1 O. 13, 0.281 146.5 (p < 0.00 1 ) -- -- -

respectively, and 24% and 3 1% for rnid and hiph dose ChEI respectively. Heterogeneity kvas

seen in almost every analysis by dosage (p4.001), except placebo and rnid dose cognitive

analysis (p=0.18. 0.1 1), low dose global assessment (p=0.31), and low dose dropouts due to

adverse events (p=0.16).

.4mi~sis by Ciciss

The ChEIs were also grouped according to first and second generation ChEIs. The analysis

revealed that both classes of medication have similar cognitive and global eficac y (289th to 32" ,,

and 270h to 29%, respectively) (Table 7 and 8). In terms of adverse evenrs. the 2 groups aiso

showed comparable rates (73% to 67%) (Table 9). However, for dropouts and dropouts due to

adverse events, second generation ChEIs had significantly smaller proportions of dropouts than

first generation ChEIs: 19% [95% CI: 16%. 23%] vs. 38% [95?h CI: 30%, Jjo6], respectively.

and 1 1 % [95% CI: 8%, 13%] vs. 27% [95% CI: 2 1 % to 32%], respectively. Heterogeneity kvas

also seen in every analysis by class (p<0.001) except for both efficacy analysis for first

generation ChEIs (p=0.5 1.0.35).

Quality Assessmen t

The quality of each study was assessed by each of the 3 expert raten. Weighted inter-rater

agreement ranged from 73% to 81% and overall agreement between raters was ~ = 0 . 2 7

(p=0.003). Seven trials received a rating of "good" (rnedian score 3), 9 trials were rated as

"satisfactory" (median score 2), and 8 trials were rated as "poor" (rnedian score 1 ). Removing

trials that were rated as poor quality studies (median quality score = 1 ) did not change the results

sipificantly (Table 6). At most a 1% to 2% increase in the difference between ChEI and

placebo was seen. Of note, after the removal of the poor quality studies, the studies included in

the adverse event analysis were homogeneous (443.9, p=0.17). The confidence intervals were

also narrowed by 5%: (Le. [95% CI: 6%, 1 1 %] compared to before [95% CI: 2%. 1 Z?'o]). In the

other cases, confidence intervals were slightly increased by 1% to 3%. Quality-weighted mean

differences were similar to analyses with al1 the studies, as well. However. the confidence

inteivals were widened by 7% to 12%. For adverse events, the quality weighted average

riifference bec2.m non-sip-ificant s the confidence internls crossed zcrc (Tabk 6).

The results O F this meta-analysis confirm that AD patients receiving treatment with oral ChEIs

demonstrate statistically significant improvements in cognition and global assessment over

placebo. The cognitive efficacy rate was 31% [95% CI: 25%. 38%] and the global assessrnent

rfficacy rate was 27% [95% CI: 23%, 31%]; however, the therapeutic effects were only lZU~, ,

[9 j% CI: 7%, 17%] and 9% [95% CI: 7%, 12%], respectively. This is consistent with previous

reviews that descnbe the benefits of ChEI as modest (13, 140-142). The proportion of

responders for placebo was 18% [95% CI: 16%, 2 1 %] for cognitive outcomes and 1 7 O h [9joa CI:

14%, 19%] for global assessment outcomes. This is similar to that found in randomized

controlled trials and underscores the benefits of non-phamacological interventions associated

with participation in research studies.

The safety analysis also detected differences between ChEI therapy and placebo. The Frequency

of treatment-emergent adverse events observed in the ChEl group was 69% compared to 6 1% for

placebo. The pooled mean difference between the two groups was statistically significant. but

the magnitude was marginal (7%, [95% CI: 2%, 12%]). The placebo rate may appear

abnomally high. One study (263) observed that 100% of subjects receiving placebo ( ~ 9 3 )

experienced an adverse event during the study. However, the types of side effects ranged From

gastrointestinal disturbances to headaches and insomnia and were described as mild and

transient. This emphasizes the frequency of minor illnesses that occur in the elderly. An

examination of dropouts may be a better representation of tolerability as higher rates ofdropouts

and dropouts due to adverse events were seen with ChEIs compared to placebo. More than one

quarter (270%) of all sübjccts md~mizcd to ChEI trcatirient did iwi coriipleie ille trial coniparred

to 15% for placebo. The pooled mean difference was 10% [95% CI: 7%. 1 4 O h I . For dropouts

due to adverse events the difference between groups was 1 1% [95% CI: 6% to 16%].

Single arm analyses were also performed to assess for the presence of a significant dose rffeci.

Cornparisons between low, mid, and high dose ChEI with placebo revealed that for cognitive

efficacy (11% vs. 18%), frequency of adverse events (62% vs. 62%), and dropout rates ( 15% vs.

15% and 6% vs. 6%) are remarkably similar between low dose ChEI and placebo. Efficacy by

global assessrnent was similar between all 3 doses of ChEI (25% to 28%), which was hiyher than

placebo (17%). For cognitive efficacy, mid dose ChEI had a larger proportion of responders

compared to high dose ChEI (35% vs. 31%). Studies have demonstrated that the response to

ChEI pharmacotherapy correlate well with red penpheral AChE inhibition (15. 270). Howcvrr.

these medications follow the standard dose-response relationship such that as the dose is

increased, the degree of AChE inhibition plateaus, as does the response. The doses classified as

mid and high for ChEIs in this study are known to produce near maximal AChE inhibition and

thus have a similar therapeutic effect. However, though this was not was seen, the high dose

ChEI group was expected to be more efficacious than mid dose. The response of the mid dose

ChEI goup was not significantly greater than the high dose group [95% Us: 31%. 40% for mid

dose vs. 23%, 39% for high dose]. But, the trend observed may be accounted for by examining

the safety analyses. High dose ChEI demonstrated greater rates of treatment-emergent adverse

events (78% vs. 65%), dropouts (37% vs. 24%), and dropouts due to adverse events (23?/0 vs.

ljOh). This is important to consider as cognitive efficacy rates reported were based on the

proportion of subjects demonstrating significant improvement. defined as a 4-point decrease in

ADAS-cog score, in an intention-to-treat (ITT) analyses. Subjects were included in these

analyses if they had received one dose of the study dmg and had one post-baseline data point.

Noticeable responses to CiiEls are not usuaily seen untii afier approximateiy i l weeks ot'

pharmacotherapy ( 15. 21, 22, 105). (This was the reason the minimum duration for studies to be

included in the meta-analysis was 12 weeks.) As a large portion of dropouts occur during the

first 4 to 8 weeks of treatment, many subjects included in ITT analyses likely only reccived a

shon t d of the ChEI and in fact dropped out of the study before a significant response could be

achieved (1 09). Unfonunately, the proportion of subjects Ming to tolerate the medication rather

than Failing to respond significantly could not be extracted. For global assessment outcornes. for

which no dose effect was observed, rates of response were based on end of treatment scores and

did not include dropouts. Thus the response of ail subjects being evaluated were based only on

completors who received a full duration of treatment.

Cornparisons were also performed between fint generation (tacrine. velnacrine. and

physostipine) and second generation (donepezil, rivastigrnine, galantamine, eptastigrnine. and

metrifonate) ChEIs. As expected the eficacy rates were similar between the two groups for both

cognition and global assessment (27% to 32%). The fiequency of adverse events was also

similar (73% to 67%); however, dropout rates were significantly different. The dropout rate for

fint generation ChEIs was 38% [95% CI: 30%, 45%] compared to 19% [95% CI: 16%, 23%] for

second generation ChEIs. For dropouts due to adverse events were at l e s t 62% and as much as

400% more likely for first generation ChEIs (27% [95% CI: 21%, 32%] vs. 11% [95% CI: 8%.

13%]. These results CO& reviews descnbing equal efficacy between first and second

generation ChEIs, but significantly better tolerability of the second generation medications ( 12.

13, 18).

The single arrn analyses undencore the importance of safety in ChEI pharmacotherapy. Though

similar benefits are seen with al1 ChEIs, donepezil. but not tacrine, was submitted for approval

for treatment of AD in Canada. And even with second generation ChEIs. tolerability may lirnit

the therapeutic response. In response, strategies such as slow titration to higher doses and

careh1 monitoring of adverse events have been implemented to maximize the potential of thesr

medications to reverse or arneliorate the cognitive decline associated with AD.

For the quality analysis, inter-rater agreement ranged kom 73% to 8 1% and was classi ficd as

*good". Overall agreement between n t e n was "fair" and significant (K = 0.27. p=0.003). Ali 24

studies received the same quality score fiorn at leasi 2 of the 3 raten. In a situation with 3 raters.

as long as 2 of the 3 agree, the mode is also the median. For exarnple. the median rating for a

study receiving a quality score of 3 fiom rater #l and rater #2 will be 3 regardless of the nting

assigned by rater #3 (e.g. the median for each of the following data sets is 3: (0.3.3 1 . 1 1.3.3 1 .

(2,3,3), {3,3,3)). Al1 3 raten assigned the s m e quality rating to 7 of the 24 studies. thus

agreement behveen pairs of raters was 313. However, for the other 17 studies, only 2 out of the 3

raters agreed, thus agreement between pairs of raters was 113 (e-g. if rater #1 agreed with rater

$2, and rater #1 did not agree with rater #3, then rater #2 also did not agree with rater #3). This

resulted in the low kappa value but had no effect on the results of the two quality analyses as the

disagreement between raten had no efiect on the final rating each study received.

Overall, the results of this meta-analysis indicate that oral ChEI pharmacotherapy in AD is, at

best, modestly efficacious cornpared to placebo, with a therapeutic effect of approximately 10%.

The tolerability of these medications is also an important consideration when evaluating their

benefits. It also appears that many AD patients may be undergoing costly treatments with no

benefit as their functioning continues to decline irreversibly. This is a major cause for concem

and hrther supports the need for methods of predicting treatment response to ChEt. The

realization of this research goal would result in more efficient and individualized selection of

treatment options.

CONCLL~SIONS

1. The cognitive therapeutic effect of oral ChEI treatment in AD is 12% [95% CI: 7h0, 17°~o].

The therapeutic effect for global assessrnent was 9% [95% CI: 7%. 17%]. The frequency of

adverse in events is 7% [95% CI: 2%. L2%] greater with ChEIs than placebo. The pooled

mean difference in dropout and dropouts due to adverse events was 10% [95% CI : 7% l A O j ]

and 1 1% [95% CI: 6%. 16%], respectively.

2. The dosing strategy for ChEIs has an important role in the tolerability and thus the rfficacy

of these medications as lengthy treatment is oRen required to achieve maximal benefits.

Similar efficacy was seen in mid (35% [95% CI: 3 1%. 40%]) and high dose (3 196 [9j0& CI:

23%, 39061) ChEIs in [TT analyses, but a larger but non-signiticant proportion of dropouts

was seen in the high dose ChEI group (37% [95% CI: 28%. 46%]) compared to mid dose

ChEI (21% [95% CI: 17%, 30%]).

3. Fint and second generation ChEIs are equally efficacious (proportion of responders ZSU'a

[95% CI: X%, 31%] vs. 32% [95% CI: 22%, 42941, respectively). However. second

pneration agents are considerably better tolerated (e.5 dropout rate 38?b [9j00 CI: 30%.

4j%] for tint generation ChEIs compared to 19% [9S0h CI: 16%. 23%] for second

generation ChEIs.

4. These findings emphasize the need for better treatment strategies and predictors of treatment

response in order to optimize ChEI pharmacotherapy in AD.

As time goes by. it seems fewer studies involving the use of pupillometq are being published.

.4fier the initial flurry of papen in the mid to late 90's. the growth is this area of the literature has

subsided. Most wiil now concede that this procedure has failed as a diagnostic test for AD.

However, there is substantial support for a link between the central and peripheral cholinergic

systems. Unfortunately, the results fiom this thesis were not able to contribute substantially to

this field by supporting or refuting this relationship as they lacked significant statistical power.

However, this work has brought to the forefront many of the limitations associated with the

methodologies employed in ihis line of research and perhaps the need to take a step back and

optimize these procedures to ensure that they both valid and reliable before further studies occur.

Clearly, the attractiveness and necessity of a simple method to diapnosis, monitor. and/or predict

response to treatments for AD increases each day as the prevalence of this disease continues to

rise. To overcome the statistical limitations of this work, studies with larger sample sizes are

necessary. Though challenges associated with recruitment may remain. part of the solution may

lie in expanding searches to include the cornmunity at large. The community group formed for

this work was only in its infancy by study's end, but showed significant potential for growth inio

functional network of physicians providing access to a relatively untapped wellspnn_r of

subjects. However, the future in this line of research must begin with the innovation of new

methodologies of ocular drug delivery and novel techniques to measure both pupillary responses

and assess central cholinergic function. For example, some studies have exarnined the use of an

ocular spray that can be adrninistered to individuals with their eyes closed. Developments to

improve the reliability, accwacy, and precision of pupillometer technology would also be

beneficial. Scales focusing exclusively on cholinergic components of cognition may also serve

as better instruments to detect changes in central cholinergic function. Lastly, inter-individual

variably between subjects must also be carefully controlled or accounted for in any prospective

studies comparing results between groups or individuais.

REFERENCES

Tomlinson BE, Blessed G, Roth M: Observations on the brains of demented old people. J Neurol Sci 1970; 1 1 (3):205-42

Canadian study of health and aging: study methods and prevalence of dementia. Cmaj 1994: 150(6):899-9 13

Evans DA, Funkenstein HH, Albert MS, Schen PA, Cook NR, Chown MJ, Hebrrt LE. Hemekens CH, Taylor JO: Prevalence of Alzheimer's disease in a community popiilation nT

older persons. Higher than previously reported [see cornments]. Jama 1989; 262( 1 8 ) : X 1-6

Wolf BA, Wertkin AM, Jolly YC, Yasuda RP, Wolfe BB, Konrad RJ, Manning D. Ravi S. Williamson IR, Lee VM: Muscarhic regulation of Alzheimer's disease amyloid preciirsor protein secretion and amyloid beta-protein production in human neuronal NT2N cells. J Biol Chem 1995; 2ïO(9):49 16-22

Eggert A, Crismon M, Erechefsky L: in Pharmacotherapy. New York, Elsevier. 1997

Terry RD, Davies P: Dementia of the Alzheimer type. AMU Rev Neurosci 1980; 3 : 77-9 j

Francis P, Cross A, Bowen D: Neurotransmitters and neuropeptides., in Alzheimer Diseasc. Edited by Terry R, Katzman R, Bick K. New York, Raven Press. 1994. pp 247-26 1.

.*aujo DM, Lapchak PA, Robitaille Y, Gauthier S, Quirion R: Differential alteration of various cholinergic markers in cortical and subcortical regions of human brain in Alzheimer's disease. J Neurochem 1988; 50(6): 19 14-23

Sims NR, Bowen DM, Allen SJ, Smith CC, Neary D, Thomas DJ, Davison AN: Presynapric cholinergic dyshction in patients with dementia. J Neurochem 1 983; 40(2):503-9

10. Wisniewski KE, Wisniewski HM, Wen GY: Occurrence of neuropathological changes and dernentia of Alzheimer's disease in Down's syndrome. Ann Neurol 1985; l7(3):?78-82

11. Giacobini E: From molecular structure to Alzheimer therapy. Jpn J Pharmacol 1997; 74(3):2X-J 1

12. Cummings JL: Cholinesterase inhibiton: A new class of psychotropic compounds. h .! Psychiatry 2000; l57(1):4- 1 5

13. Mayeux R, Sano M: Treatment of Alzheimer's disease. N Engl J Med 1999; 341(22):1670-9

14. Barner EL, Gray SL: Donepezil use in Alzheimer disease. Ann Pharmacother 199s: 32(1):70-7

15. Rogers SL, Friedhoff LT: The efficacy and safety of donepezil in patients with Alzheimer's disease: results of a US Muhicentre, Randomized, Double-Blind, Placebo- Controlled Trial. The Donepezil Study Group. Dementia 1996; 7(6):293-303

16. Peters BH, Levin HS: Effects of physostigmine and tecithin on memory in Alzheimer disease. Ann Neurol 1 979; 6(3):2 1 9-2 1

17. Qizilbash N, Whitehead A, Higgins J, Wilcock G, Schneider L, Farlow M: Cholinesterase inhibition for Alzheimer disease: a meta-analysis of the tacrine tials. Dementia Trialists' Collaboration. Jarna 1998; 280(20): 1777-82

18. McGleenon BM, Dynan KB, Passmore AP: Acetylcholinesterase inhibitors in Alzheimer's disease. Br J Clin Pharmacol 1999; 48(4):471-80

19. Knapp MJ, Knopman DS, Solomon PR, Pendlebury WW, Davis CS, Gracon SI: A 30-week randomized controlled trial of high-dose tacnne in patients with Alzheimer's disease. The Tacrine Study Group. Jama 1994: 27 1 ( 13):985-9 1

20. Farlow M. Gracon SI, Hershey LA, Lewis KW, Sadowsky CH, Dolan-Ureno J: A controlled trial of tacrine in Alzheimer's disease. The Tacnne Study Group Jarna 1992: 268( 1 8):2523-9

21. Rogers SL, Farlow MR, Doody RS, Mohs R, Friedhoff LT: A 24-week. double-blind. placebo-controlled trial of donepezil in patients with Alzheimer's disease. Donepezil Study Group. Neurology 1998; 50(1): 136-45

22. Rogers SL, Doody RS, Mohs RC, Friedhoff LT: Donepezil improves cognition and global function in Alzheimer disease: a 1 Sweek, double-blind, placebo-controlled study. Donepezil Study Group. Arch Intem Med 19%; l58(9): 102 1-3 1

23. Scinto LF, Daffher KR, Dressler D, Ransil BI, Rentz D, Weintraub S, Mesularn M. Potter H: A potential noninvasive neurobiological test for Alzheimer's disease Science 1994: 266(5l87): 105 1-4

24. Kardon RH: Drop the Alzheimer's drop test. Neurology 1998; 50(3):588-91

25. Naranjo C, Fourie J, Hernnann N, Lanctot K, Birt C, Yau K: Probing central and peripherai cholinergic responses. J Psyc hiatry Neurosci 2000; X(4):3 25-336.

26. Berger-Sweeney J, Heckers S, Mesulam MM, Wiley RG, Lappi DA, Shama M: Differential effects on spatial navigation of irnmunotoxin-induced cholinergic lesions of the medial septal area and nucleus basalis magnocellularis. I Neurosci 1994; l4(7):&07- 19

27. Muir JL, Everitt BJ, Robbins TW: AMPA-induced excitotoxic lesions of the basal forebrain: a significant role for the cortical cholinergic system in attentional hinction. J Neurosci 1994: 14(4):23 13-36

28. Voytko ML, Olton DS, Richardson RT, Gorman LK, Tobin SR, Pnce DL: Basal forebrain Iesions in monkeys disrupt attention but not learning and memory. J Neurosci 1994: 14(1): 167-86

29. Dunnett SB: The role and repair of forebrain cholinergic systems in short-term memory. Studies using the delayed matching-to-position task in rats. Adv Neurol 1993; 5953-65

30. Squire LR, Knowlton B, Musen G: The structure and organization of memory. h u Rev Psycho1 1993; 44453-95

3 1. Broks P, Preston GC, Traub M, Poppleton P, Ward C, Stahl SM: Modelling dementia: effects of scopolamine on memory and attention. Neuropsychologia 1988; 26(5):685-700

32. Sherman KA, Kuster JE, Dean RL, Bartus RT, Friedman E: Prespaptic cholinergic mechanisms in brain of aged rats with rnemory impairments. Neurobiol Aging 1 98 1 ; 2(1):99- 1 O4

33. Bamis RT, Dean Ri,, Pontecorvo MJ, Flicker C: The cholinergic hypothesis: a historical overview, current perspective, and future directions. Ann N Y Acad Sci 1985; LWJ:332-58

34. Drachman DA, Leavitt J: Human memory and the cholinergic system. A relationship to aging? Arch Neurol 1974; 3O(2): 1 13-2 1

35. Black S: Alzheimer's: update on new dnig treatrnents. Can J CME l994( 1 1 ):69-82

36. Davis EU, Mohs RC, Tinklenberg JR, Pfefferbaum A, Hollister LE, Kopell BS: Physostigmine: improvement of long-term memory processes in normal humans. Science 1978; 20 l(4352):272-4

37. Flicker C, Fems SH, Serby M: Hypersensitivity to scopolamine in the rlderly. Psychopharmacology 1992; 1 O7(2-3): 437-4 1

38. Wesnes K, Anand R, Lorscheid T: Potential of moclobemide to improve cerebral insufficiency identified using a scopolamine model of aging and dementia. Acta Psychiatr Scand Suppl 1990; 360:71-2

39. Vannucchi MG, Scali C, Kopf SR, Pepeu G, Casamenti F: Selective muscarinic antagonists differentially affect in vivo acetylcholine release and mernory performances of young and aged rats. Neuroscience 1 997; 79(3):837-46

40. Aubert 1, Rowe W, Meaney MJ, Gauthier S, Quirion R: Cholinergic markers in aged cognitively impaired Long-Evans rats. Neuroscience 1995; 67(2):277-92

41. Araujo DM, Lapchak PA, Meaney MJ, Collier B, Quirion R: Effects of aging on nicotinic and muscarinic autoreceptor huiction in the rat brain: relationship to presynaptic cholinergic markers and binding sites. J Neurosci 1990; 1 0(9):3069-78

42. Dubois B, Pilon B, Lhermitte F, Agid Y: Cholinergic deficiency and frontal dysfunction in Parkinson's disease. Ann Neurol 1990; 28(2): 1 1 7-2 1

43. Kopelman MD, Corn TH: Cholinergic 'blockade' as a model for cholinergic depletion. .\ cornparison of the memory deficits with those of Alzheimer-type dernentia and the alcoholic Konakoff syndrome. Brain 1988; 1 1 1(Pt 5): 1079-1 10

44. Witt ED: Neuroanatomicai consequences of thiamine deficiency: a comparative analysis. Alcohol Alcohol 1985; 20(2):201-21

45. Katzung B: Basic & Clinical Pharmacology. East Norwaik, CT., Appleton & Lange, 1999

16. Kalant H, Roschlan W: Principles of Medical Pharmacology. New York, Oxford University Press, 1998

47. Guyton A: Textbook of Medical Physiology. Philadelphia, W.B. Saunden Company, 1991

48. Borner TI: The molecular basis of muscarinic receptor diversity. Trends Neurosci 1989; l2(4): 148-5 1

49. Offermanns S, Wieland T, Homann D, Sandmann J, Bombien E. Spicher K. Schultz G. Jakobs KH: Transfected muscarinic acetylcholine receptors selectively couple to Gi- type G proteins and Gq/ 11. Mol Pharmacol 1994; 45(5):890-8

50. Levey AI, Kitt CA, Simonds WF, Price DL, Brann MR: Identification and localization of muscarinic acetylcholine receptor proteins in brain with subtype-specific antibodies. J Neurosci 199 1; 1 l(1 O):3218-26

5 1. Levey AI: Muscarinic acetylcholine receptor expression in memory circuits: implications For treatment of Alzheimer disease. Proc Nat1 Acad Sci U S A 1996; 93(24): 13541-6

52. Flynn DD, Ferrari-DiLeo G, Mash DC, Levey AI: Differeritial regulation of molecular subtypes of muscarhic receptors in Alzheimer's disease. J Neurochem 1995; 64(4): 1 888-9 1

53. Evans D: Descriptive epiderniology of Alzheimer's disease., in Alzheimer's disease. Cause(s), diagnosis, treatment and care. Edited by Khachatunan Z, Radebaugh T. New York. CRC Press, 1996, pp 5 1-59.

54. Sekoe DJ: Aging brain, aging mind. Sci Am 1992; 267(3): 134-43

55. Cummings J, Benson D: Dementia: a clinical approach. Boston, Heinemann-Buttenvorth. 1992

56. Jann MW:Preclinical pharmacology of metrifonate. Pharmacotherapy 1998; l8(2 Pt 2):jj-67

57. Pavia J, de Ceballos ML, Sanchez de la Cuesta F: Muscarinic receptors in Alzheimer's disease. Methods Find Exp Clin Pharmacol 1996; 1 I(Supp1 B): 7 1-5

58. Rocca WA, Hohan A. Brayne C, Breteler MM, Clarke M, Copeland JR, Dartigues JF. Engedal K. Hagnell O, Heeren TJ, et al.: Frequency and distribution of Alzheimer's disease in Europe: a collaborative study of 1980- 1990 prevalence findings. The EüRODEM- Prevalence Research Group. Ann Neurol 199 1 ; 30(3):38 1-90

59. Wenk GL: The nucleus basalis magnocellularis cholinergic system: one hundred years of progress. Neurobiol Leam Mem 1997; 67(2):85-95

60. Fowler CJ, O'Neill C, Winblad B, Cowbum RF: Neurotransmitter, receptor and signal transduction disturbances in Alzheimer's disease. Acta Neurol Scand Suppl 1992; l39:59-6?

61. Oyanagi K, Takahshi H, Wakabayashi K, h t a F: Correlative decrease of large neurons in the neostriatum and basal nucleus of Meynert in Alzheimer's disease. Brain Res 1989; 504(2):354-7

62. Cummings L: Metrifonate: overview of safety and efficacy. Pharmacotherapy 1998; 1 8(2 Pt 2):43-6

63. Wenk GL, Stoehr JD, Quintana G, Mobley S, Wiley RG: Behavioral, biochemical, histological, and electrophysiological effects of 192 IgG-saporin injections into the basal forebrain of rats. J Neurosci 1994; 14(10):5986-95

64. Dunnett SB: Comparative effects of cholinergic drugs and lesions of nucleus basalis or fimbna-fornix on delayed matching in rats. Psychopharmacology 1985; 87(3):357-63

65. Flicker C. Dean R, Bartus RT, Fems SH, Crook T: Animal and hurnan memory dysfunctions associated with aging, cholinergic lesions, and senile dementia. Am N Y Acad Sci 1985; 4445 15-7

66. Bierer LM, Haroutunian V, Gabriel S, Knott PJ, Carlin LS, Purohit DP, Peri DP. Schmeidler J, Kanof P, Davis KL: Neurochemical correlates of dementia seventy in Alzheimer's disease: relative importance of the cholinergic deficits. J Neurochem 1995; 64(2):749-60

67. Lehericy S, Hirsch EC, Cervera-Pierot P. Hersh LB. Bakchine S, Piette F. Duyckaens C. Hauw JJ, Javoy-Agid F, Agid Y: Heterogeneity and selectivity of the degeneration of cholinergic neurons in the basal forebrain of patients with Alzheimer's disease. J Comp Neurol 1993; 330(1): 15-3 t

68. Doucette R, Fisman M, Hachinski VC, Mersky H: Cell loss from the nucleus basalis of Meynert in Alzheimer's disease. Can J Neurol Sci 1986; 13(4 Suppl):435-40

69. Peny EK, Tomlinson BE, Blessed G, Bergrnann K, Gibson PH, Perry RH: Correlation of cholinergic abnormalities with senile plaques and mental test scores in senile dementia. Br Med J 1978; 2(6 1 50): 1457-9

70. Sunderland T. Tariot PN, Cohen RM, Weingartner H. Mueller EAd. Murphy DL: Anticholinergic sensitivity in patients with dementia of the Aizheirner type and age-matched controls. A dose-response study. Arch Gen Psychiatry 1987; JJ(5):118-26

71. Rabey JM, Neufeld MY, Treves TA, Sifris P, Korczyn AD: Cognitive effects of scopolamine in dementia. J Neural Transm Gen Sect 1996; 103(7):873-8 1

72. Pomara N, Nolan K, Halpern G: Scopolamine-induced impairment as a potential predictor ol' Alzheimer's disease in individuals with Apolipoprotein E type 4 alleles. Neurochem Res 1995; 20(12): 1 5 19-20

73. Rodriguez-Puertas R, Pascual J, Vilaro T, Pazos A: Autoradiographic distribution of M l . M2, M3, and M4 muscarinic receptor subtypes in Alzheimer's disease. Synapse 1997: 26(4):341-50

74. Kopelman MD: The cholinergic neurotransmitter system in hurnan memory and dementia: a review. Q I Exp Psycho1 [A] 1986; 38(4):535-73

75. H e m m N, Lanctot KL: From transmitten to treaûnent: the phannacothenpy of behavioural disturbances in dementia. Can J Psychiatry 1997; 12 Suppl 1 :5 1 S-64s

76. Reinikainen KJ, Soininen H, Riekkinen PJ: Neurotransmitter changes in Alzheimer's disease: implications to diagnostics and therapy. J Neurosci Res 1990; 27(4):576-86

77. Newell F: Ophthalmology: Principles and Concepts. St. Louis, Mosby-Year Books, Inc., 1996

78. Thompson H: The Pupil. in Adler's Physiology of the Eye. Edited by Hart W, Jr. St. Louis. Mosby-Year Books, Inc., 1992, pp 412-441

79. Gupta N, McAliister R Drance SM, Rootman J, Cynader MS: Muscarhic receptor Ml and M2 çubtypes in the hurnan eye: QNB, pirenzipine, oxotremorine, and AFDx--1 16 in vitro autoradiography. Br J ûphthahol 1994; 78(7):555-9

Gupta N, Drance SM, McAllister R, Prasad S, Rootman J, Cynader MS: Localization of M3 muscarinic receptor subtype and rnRNA in the human eye. ûphthalrnic Res 1994; 26(4):207- 13

8 1. Bloom F: in Goodman and Gilman's: The pharmacological basis of therapeutics. Edited by Hardman J, Limbird L, Molinoff P, Ruddon R Gilman A. Toronto, Pergamon Press. 1998

82. Appleyard ME, McDonald B: Reduced adrenal gland acetylcholinesterase activity in Alzheimer's disease. Lancet 199 t ; 338(8774): 1085-6

83. Sacks B, Smith S: People with Down's syndrome cm be distinguished on the basis of cholinergie dysfunction. J Neurol Neurosurg Psychiatry 1989; 52(11): 1294-5

84. Hams WS, Goodman RM: Hyper-reactivity to atropine in Down's syndrome. N Enpl J bled i968; 279(8):407-10

85. Berg J, Gillian-Brandon M, Kriman B: Atropine in mogolism. Lancet 1959; ii:U.ll-2.

86. Scinto LF, Wu CK, Firla KM, Damier KR, Saroff D, Geula C: Focal pathology in the Edinger-Westphal nucleus explains pupillary hypersensitivity in Alzheimer's disease. .Acta Neuropathol (Berl) 1999; 97(6):557-64

87. Growdon JH, Graefe K, Tennis MT Hayden D. Schoenfeld D. Wray SH: Pupil dilation to tropicarnide is not specific for Alzheimer disease. Arch Neurol 1997; j4(7):83 1-4

88. Loupe DN, Newman NI, Green RC, Lynn MI, KK WI, Geis TC. Edelhauser HF: Pupillary response to tropicamide in patients with Alzheimer disease. Ophthalmology 1996: 1 O3(3):495-503

89. Fridh MT Havelius U, Elofsson G, Hindfelt B: The pupillary response to tropicamide in Alzheimer's disease. Acta Op hthalrnol Scand 1996; 74(3):276-9

90. Kaneyuki H, Mitsuno S, Nishida T, Yarnada M: Enhanced miotic response to topical dilute pilocarpine in patients with Alzheimer's disease. Neurology 1998; 50(3):802-4

9 1. Katz B: Detecting Alzheimer's disease. Science 1995; 267(5204): 1578-79

92. Hannannel M, Felier Ofiy V, Villagra R, Martin R: Parasympathetic function of the eue in dementia of the Alzheimer type @AT). Neurology 1 995; 45 :A3 56.

93. Idiaquez J, Alvarez G, Villagra R, San Martin RA: Chotinergic supersensitivity of the iris in Alzheimer's disease [letter]. J Neurol Neurosurg Psychiatry 1994; jT(l2): 1 544-5

94. Pomara N, Sitaram N: Detecting Alzheimer's disease. Science 1995; 267(5204):1579-80;

95. Davis KL, Mohs RC: Enhancement of memory processes in Alzheimer's disease with multiple- dose intravenous physostigmine. Am J Psychiatry 1982; 139(11): 142 1-4

96. Thal LJ, Fuld PA, Masur DM, Sharpless NS: Oral physostigmine and lecithin improve memory in Alzheimer disease. Ann Neurol 1983; 13(5):49 1-6

97. Smith CM, Swash M. Exton-Smith AN, Phillips MJ. Overstall PW, Piper ME, Bailey MR: Choline therapy in Alzheimer's disease [letter]. Lancet 1978; 2(8084):3 18

98. Fovall PT Dysken MW, Lazarus LW, Davis IM, Kahn RL, Jope R, Finkel S, Rattan P: Choline bitartrate treatment of Alzheimer-type dementias. Commun Psychopharmacol 1980: 4(2): 1 4 1 -5

99. Renvoize EB, Jenam T: Choline in Alzheimer's disease. N Engl J Med 1979; 30 1 (6):33O

100. Thal LJ, Rosen W, Sharpless NS, Crystal H: Choline chloride fails to improve cognition of Alzheimer's disease. Neurobiol Aging 198 1 ; 2(3):205-8

101. Brinhan SD, Smith RC, Meyer JST Vroulis G. Shaw T. Gordon IR, Allen RH: Lrcithin and memory training in suspected Alzheimer's disease. J Gerontol 1982; 37( 1 ):+9

102. Dysken MW, Fovall P, Harris CM, Davis JM, Noronha A: Lecithin administration in Alzheimer dementia [letter]. Neurology 1982; 32( 10): 1203-4

103. Little A, Levy R, Chuaqui-Kidd P, Hand D: A double-blind, placebo controlled trial of high-dose lecithin in Alzheimer's disease. J Neurol Neurosurg Psychiatry 1985; 48(S):736-42

104. Weintraub S, Mesulan MMT Auty R Baratz R, Cholakos BN. Kapust L, Rand B. Teilen JG, Albert MS, LoCastro S, Moss M: Lecithin in the treatment of Alzheimer's disease [letter]. Arch Neurol 1983; 10(8):527-8

105. Burns A, Rossor M. Hecker J, Gauthier S, Petit H, Moller HI, Rogers SL, Friedhoff LT: The effects of donepezil in Alzheimer's disease - results from a multinational trial. Dement Geriatr Cogn Disord 1999; 10(3):237-44

106. Davis EU, Thal LJ, Garnzu ER, Davis CS, Woolson RF, Gracon SI, Drachman DA. Schneider LS, Whitehouse PJ, Hoover TM, et al.: A double-blind, placebo-controlled multicenter study of tacrine for Alzheimer's disease. The Tacrine Collaborative Study Group [see comments]. N Engl I Med 1992; 327(l8): 1253-9

107. Tariot PN: Evaluating response to rnetrifonate. J Clin Psychiatry 1998; 59(Suppl9):3 3-7

108. Schwartz D: Self-reported illness and health status among Gulf War veterans. Journal of the Amencan Medical Association 2 997; 277:23 8-45.

109. Giacobini E: lnvited review: Cholinesterase inhibitors for Alzheimer's disease therapy: from tacnne to future applications. Neurochem Int 1998; 32(5-6):413-9

1 10. Drachman D, Resiberg B (eds): Alzheimer's disease: the standard reference. New York. The Free Press, 1983

1 1 1. Karczrnar A: Invited review : Anticholinesterases: dramatic aspects of their use and misuse. Neurochem Lnt 19%; 32(5-6):401-11

1 12. Liberini P. Valerio AT Memo M, Spano PF: Lewy-body dementia and responsiveness to cholinesterase inhibitors: a paradigm for heterogeneity of Alzheimer's disease? Trends Phannacol Sci 1996; 17(4): 155-60

113. Association AP: Diagnostic and Statistical Manual of Mental Disorders. Washington D.C.. Amencan Psychiatnc Association, 1994

114. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan Ebl: Ciinical diagnosis of Alzheimer's disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer's Disease. Neurology 1984; 34(7):939-44

115. Moroi S, Lichter P: in Goodman and Gilman's: The phmacological basis of therapeutics. Edited by Hardman I, Limbird L, Molinoff PT Ruddon R, Gilman A. Toronto. Pergarnon Press, 1998

116. Weintraub M. Evans P: Pilocarpine: the method is the message. Hospital Formula- 1985; 20:177-181.

1 17. Kaplan H: Rapid computerized determination of pupil diameter fiom imperfect video images. Rockville. MD. National Institute on Drug Abuse. 1995

1 [S. Rosen WG, Mohs RC, Davis KL: A new rating scale for Alzheimer's disease. . lm J Psychiatry 1984; 1 1 1 ( 1 1 ): 1356-64

119. Folstein MFT Folstein SE, McHugh PR: "Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. .J Psychiatr Res 1975; 12(3): 189-95

120. Gauthier S, Gelinas 1, Gauthier L: Functional disability in Alzheimer's disease. Int Psychogenatr 1997; 9(Suppl 1): 163-5

121. Knopman DS, Knapp MJ, Gracon SI, Davis CS: The Clinician Interview-Based Impression (CM): a cliniciads global change rating scale in Alzheimer's disease. Neurology 1994; M(l2):23 1 5-2 1

122. Cognex: Expanded access proposed by FDA as part of "program of further study": Warner-Lambert will NDA studies with other data by Apnl 1. F-D-C Rep 199 1

113. Talwalker S, Overall JE, Srirama MK, Gracon SI: Cardinal features of cognitive dysfunction in Alzheimer's disease: a factor-analytic study of the Alzheimer's Disease .Assessrnent Scale. J Geriatr Psychiatry Neurol 1996; 9(1):39-46

124. Rogers SL, Friedhoff LT: Long-tenn efficacy and safety of donepezil in the treatment of Alzheimer's disease: an interim analysis of the resuits of a US multicentre open label extension study [see comments]. Eur Neuropsychopharmacol 1998; 8(1):67-75

125. Weisgraber KH, Roses AD, Strittmatter WJ: The role of apolipoprotein E in the nervous system. Curr Opin Lipidol 1994; 5(2): 1 10-6

126. Saunders AM, Stnttmatter WJ, Schmechel D, George-Hyslop PH, Pericak-Vance MA. Joo SH, Rosi BL, Gusella JF, Crapper-MacLachlan DR, Aiberts MJ. et al.: Association of apolipoprotein E allele epsilon 4 with late-onset familial and sporadic Alzheimer's disease. Neurology 1993; 43(8): 1467-72

127. Higuchi S, Matsushita S, Hasegawa Y, Muarnatsu T, Arai H, Hayashida M: Apolipoprotein E epsilon 4 allele and pupillary response to tropicamide. Am J Psychiatry 1 997; l54(5):694-6

128. Ferrario E, Molaschi M, Villa L, Varetto O, Bogetto C. Nuvi R: Is videopupillography useful in the diaenosis - of Alzheimer's disease?. Neurology 1 998; K!(?):642-4

129. Sitaram N, Pomara N: Increased pupillary miotic response to pilocarpine in cognitivrly impaired elderly subjects. WCS Medical Science 198 1; 9:4O9- 10.

130. Arai H, Terajima M, Nakagawa T, Higuchi S, Mochizuki H. Sasaki H: Pupil dilatation assay by tropicamide is modulated by apolipoprotein E epsilon 4 allele dosage in Alzheimer's disease. Neuroreport 1996; 7(4):9 1 8-20

13 1. Graff-Radford NR, Lin SC, Brazis PW, Bolling JP, Liesegang TJ, Lucas JA. Uitti RJ. O'Brien PC: Tropicamide eyedrops cannot be used for reliable diagnosis of Alzheimer's disease [see comments]. Mayo Clin Proc 1997; 72(6):495-504

132. Nagataki S, Mishima S: Pharmacokinetics of instilled drugs in the human eye. Int Ophthalmol Clin 1980; 20(3):33-49

133. Mishima S, Gasset A, Klyce SD, Jr., Baum JL: Determination of tear volume and trar flow. Invest Ophthalmol 1966; 5(3):264-76

134. FitzSimon JS, Waring SC, Kokmen E, McLaren JW, Brubaker RF: Response of the pupil to tropicamide is not a reliable test for Alzheimer disease. Arch Neurol 1997; 54(2): 155-9

135. Rosler M, Anand R, Cicin-Sain A, Gauthier S, Agid Y, Dal-Bianco P. Stahelin HB. Hartman R Gharabawi M: Efficacy and safety of rivastigmine in patients with Alzheimer's disease: international randomised controlled trial. Bmj 1999; 3 l8(7 184):633-5

136. Raskind MA, Cyrus PA, Ruzicka BB, Gulanski BI: The effects of metrifonate on the cognitive, behavioral, and functional performance of Alzheimer's disease patients. Metrifonate Study Group. J Clin Psychiatry 1999; 6O(S):3 18-75

137. Dubois B, McKeith 1, Orgogozo SM, Collins O, Meulien D: A multicentre, randomized. double-blind, placebo-controlled study to evaluate the efficacy, tolerability and safety of two doses of metrifonate in patients with mild-to-moderate Alzheimer's disease: the MALT study. Int J Genatr Psychiatry L 999; l1(11):973-82

138. Raskind MA, Peskind ER, Wessel T, Yuan W: Galantamine in AD: A 6-month randomized, placebo-controlled trial with a 6-month extension. The Galantamine USA- 1 Study Group. Neurology 2000; 54(12):226 1-8

139. Tariot PN, Solomon PR, Moms JC, Kershaw PT Lilienfeld S, Ding C: A Srnonth. randomized, placebo-controlled trial of galantamine in AD. The Galantamine USA-IO Study Group. Neurology 2000; 54( 1 2):2269-76

140. Francis PT, Palmer AM, Snape M, Wilcock GK: The cholinergic hypothesis of Alzheimer's disease:a review of progress. J Neurol Neurosurg Psychiatry 1999; 6612): 137-47

141. Krall WI, Sramek JJ, Cutler NR: Cholinesterase inhibitors: a therapeutic strategy For Alzheimer disease. Ann Phamiacother 1999; 33(4):441-50

142. Thal LJ: Cholinomirnetic treatment of Alzheimer's disease. Prog Brain Res 1996: 1 O9:299-309

143. Einarson T R Leeder IS, Koren G: A method for meta-analysis of epidemiological studies. Drug Intell Clin Pham 1988; 22(1 O):8 13-24

144. Livingston GT Katona C: How useful are cholinesterase inhibiton in the treatment ol' Alzheimer's disease? A number needed to treat analysis. Int J Geriatr Psychiatry 1000: 15(3):203-7

145. Arsland D. Laake K: [Should Alzheimer disease be treated with tacrine'? Review of the literature (see cornrnents)]. Tidsslcr Nor Laegeforen 1996; 1 l6(23):279 1-4

136. Gracon S, Smith F, Shokry A: [Advantages in long-term treatment with Cognex in Alzheimer disease]. Schweiz Rundsch Med Prax 1997; 86(35): 1350-4

147. Wettstein A: [Cholinesterase inhibitor and ginkgo extracts in therapy of dementia. A cornparison of effectiveness based on controlled studies]. Fortschr Med 1999; 1 l7(j):48-9

148. Dersirnonian R, Laird N: Meta-anaiysis in ciinical trials. Control Clin Trials 1956: 7(3): 177-88

149. Velanovich V: Meta-analysis for combining Bayesian probabilities. Med Hypotheses 1991; 35(3):192-5

150. Breslow N, Day N: The analysis of case-control studies. Statistical methods in cancer research. IARC Scientific Publication 1980; 32( 1 )

151. Laine K, Palovaara S, Tapanainen P, Manninen P: Plasma tacnne concentrations are significantly increased by concomitant hormone replacement therapy. Clin Pharmacol Ther 1999; 66(6):602-8

152. Farlow EVIR, Cyrus PA, Nadel A, Lahiri DK, Brashear A, Gulanski B: Metrifonate treatrnent of AD: influence of APOE genotype. Neurology 1999; 53(9):20 1 0-6

153. Heinig R, Sachse R: The effect of food and time of administration on the phannacoicinetic and pharmacodynamie profile of metrifonate. Lnt J Clin Pharmacol Ther 1999; 3 7(9):456-64

154. Moller HJ, Hampe1 H, Hegerl U, Schmitt W, Walter K: Double-blind, randomized. placebo-controlled clinical trial on the efficacy and tolerability of a physostigmine patch in patients with senile dementia of the Alzheimer type. Pharmacopsychiatry 1999; 32(3):99- 1 06

155. Asthana S, Raffaele KC, Greig NH, Schapiro MB, Blackman MR, Soncrant TT: Neuroendocrine responses to intravenous infusion of physostigmine in patients with Alzheimer disease. Alzheimer Dis Assoc Disord 1999; l3(2): 102-8

lj6. Steele LS, Glazier RH: 1s donepezil effective for treating Alzheimer's disease?. Can Fam Physician 1999; 6 :9l7-9

157. Srarnek JJ, Hourani J, h e e SS, Cutler NR: NXX-066 in patients with Alzheimer's disease: a bridging study. Life Sci 1999; 64(14): 12 15-2 1

158. hloss DE, Berianya P, Hasan MM, Sandovai H, ishida C: Methanesuiionyi tluondc (MSF): a double-blind, placebo-controlled study of safety and efficacy in the treatment of senile dementia of the Alzheimer type. Alzheimer Dis Assoc Disord 1 999; 1 3( 1 ):?O-5

159, Iann MW, Cyms PA, Eisner LS, Margolin DI, Griffin T, Gulanski B: Efficacy and safsty of a loading-dose regimen versus a no-loading-dose regimen of rnetrifonate in the symptomatic treatment of Alzheimer's disease: a randomized, double-masked. placebo- controlled trial. Metrifonate Study Group. Clin Ther 1999; 2 1 (l):88- 102

160. Riekkinen P, Jr., Riekkinen M: THA improves word pnrning and clonidinr enhances tluency and working memory in Alzheimer's disease. Neuropsychopharmacology 1999: 30(4):357-64

161. Imbimbo BP, Verdelli G, Martelli P, Marchesini D: Two-year treatment of Alzheimer's disease with eptastigmine. The Eptastigmine Study Group. Dement Geriatr Cogn Disord 1999; 1 O(2): 139-47

162. Roberts CJ, Ford JM, Truman CA, Scott M, Makela PM, Wilcock GK: Assessment of the value of therapeutic monitoring of tacnne in Alzheimer's disease. Eur J Clin Pharmacol 1998; 54(9-1 O):72 1-4

163. Rogers SL, Cooper NM, Sukovaty R, Pederson JE, Lee IN. Fnedhoff LT: Pharmacokinetic and pharmacodynamic profile of donepezil HC1 following multiple oral doses. Br J Clin Pharmacol 1998; 46 Suppl 1 :7-12

164. Rogers SL. Friedho ff LT: Pharmacokinetic and pharmacodynamic profile of donepezil HCI following single oral doses. Br J Clin Pharmacol 1998; 46 Suppl 1 : 1-6

165. Itil TM, Eralp E, Ahrned 1, Kunitz A, Itil KZ: The pharmacological effects of ginkgo biloba, a plant extract, on the brain of dementia patients in cornparison with tacrine. Psychopharmacol Bull 1998; 34(3):39 1-7

166. Mant T, Troetel WM, Irnbimbo BP: Maximum tolerated dose and pharmacodynamies of eptastigmine in elderly healthy volunteers. J Clin Pharmacol 1998; 38(7):6 10-7

167. McKeith IG: The clinicd trial protocol of the Metrifonate in Alzheimer's Trial (MALT). Dement Geriatr Cogn Disord 1998; 9(Suppl2):2-7

168. Pettigrew LC, Bieber F, Lettien J, Wermeling DP, Schmitt FA, Tikhtman AJ, Ashford JW, Smith CD, Wekstein DR, Markesbery WR, Orazem J, Ruzicka BB, Mas J, Gulanski B:

Pharmacokinetics, pharmacodynamies, and safety of metrifonate in patients with Alzheimer's disease. J Clin Pharmacol 1998; 38(3):236-45

169. Blin J, Ivanoiu A, De Volder A, Michel C, Bol A, Verellen C, Seron X, Duprez T. Laterre EC: Physostigmine results in an increased decrement in brain glucose consumption in Alzheimer's disease. Psychopharmacology (Berl) 1998; 1 36(3):256-63

170. Farlow MR, Lahiri DK, Poirier J, Davignon J, Schneider L, Hui SL: Treatment outcomc of tacnne therapy depends on apolipoprotein genotype and gender of the subjects with Alzheimer's disease. Neuro logy 1 998; 5O(3): 669-77

7 - v 1 1. Sc'meider LS, Fariow MR, Pogoda JM: Potenrial role for estrogen replacement in the

treatment of Alzheimer's dementia. Am J Med 1997; 103(3A):46S-50s

172. Oakley F, Sunderland T: Assessrnent of motor and process skills as a rneasure of IADL functioning in phmacologic studies of people with Alzheimer's disease: a pilot study. Int Psychogeriatr 1997; 9(2): 197-206

173. Raskind MA, Sadowsky CH, Sigmund WR, Beitler PI, Auster SB: Effect of racrine on language, praxis, and noncognitive behavioral problems in Alzheimer disease. Arch Nrurol 1 997; 54(7):836-40

174. Riekkinen P, Jr., Riekkinen M, Soininen H, Kuikka J, Laakso M. Riekkinen P. Sr.: Frontal dysfùnction blocks the therapeutic effect of THA on attention in Alzheimer's disease. Neuroreport 1997; S(8): 1845-9

175. Amar K, Wilcock GK, Scot M, Lewis T: The presence of leuko-araiosis in patients with Alzheimer's disease predicts poor tolerance to tacrine, but does not discriminate responders From non-responders. Age Ageing 1997; 26(1):25-9

176. Farlow MR, Lahin DK. Poirier I, Davignon J, Hui S: Apolipoprotein E genotype and gender influence response to tacnne therapy. Ann N Y Acad Sci 1996; 802: 10 1 - 10

177. Talwalker S: The cardinal features of cognitive and noncognitive dysfunction and the differential efficacy of tacrine in Alzheimer's disease patients. I Biopharm Stat 1996: 6(4):443-56

178. Smith F: Mixed-mode1 analysis of incornplete longitudinal data From a high-dose trial of tacrine (Cognex) in Alzheimer's patients. J Biopharm Stat 1996; 6(1):59-67

179. Smith F, Talwalker S, Gracon S, Srirarna M: The use of survival analysis techniques in evaluating the effect of long-tem tacrine (Cognex) treatment on nursing home placement and mortality in patients with Alzheimer's disease. .l Biopharm Stat 1996; 6(4):395-409

1 80. Canal N, Imbirnbo BP : Relationship between pharmacodynamic activi ty and cognitive effects of eptastigrnine in patients with Alzheimer's disease. Eptastigmine Study Group. Clin Pharmacol Ther 1996; 60(2):218-28

181. Foster NL, Petenen RC, Gracon SI, Lewis K: An e~ched-population, double-blind. placebo-controlled, crossover study of tacrine and lecithin in Alzheimer's disease. The Tacrine 970-6 Study Group. Dementia 1996; 7(5):260-6

182. Fontana RI, Turgeon DK, Woolf TF, Knapp MJ, Foster NL, Watkins PB: The caffeinr breath test does not identiS, patients susceptible to tacrine hepatotoxicity. Hepatology 1996: 23(6): 1429-35

183. Knopman D, Schneider L, Davis K, Talwalker S, Smith F, Hoover T, Gracon S: Long- term tacrine (Cognex) treatment: effects on nuning home placement and mortality. Tacrine Study Group. Neurology 1996; 47(1): 166-77

184. Harkins SW, Taylor IR, Mattay VS: Response to tacrine in patients with dementia of the Alzheimer's type: cerebral perfusion change is related to change in mental status. Int 1 Neurosci 1996; 84(1-4): 149-56

185. Zemlan FP: Velnacnne for the treatment of Alzheimer's disease: a double-blind, placebo- controlled trial. The Mentane Study Group. J Neural Transm 1 996; 1 03(8-9): 1 1 05- 1 6

186. Zemlan FP, Keys M. Richter RW, Strub RL: Double-blind placebo-controllrd study of velnacrine in Alzheimer's disease. Life Sci 1996; 58(2 1): 1523-32

1 87. Gracon SI: Evaluation of tacrine hydrochloride (Cognex) in two parallei-group studirs. Acta Neurol Scand Suppi 1996; 165: 1 14-22

188. Sramek JJ. Anand R, Wardle TS, Irwin P, Hartman RD, Cutler NR: Safety/tolerabilitv trial of SDZ ENA 713 in patients with probable Alzheimer's disease. Life Sci 1996; S8(lS): 110 1-7

189. Marin DB, Bierer LM. Lawlor BA. Ryan TM, Jacobson R, Schmeidler I. Mohs RC. Davis KL: L-deprenyl and physostigmine for the treatment of Alzheimer's disease. Psychiatry Res 1995; 58(3): 18 1-9

190. Minthon L, Nilsson K, Edvinsson L, Wendt PE. Gustafson L: Long-term effects of tacnne on regional cerebral blood flow changes in Alzheimer's disease. Dementia 1995: 6(5):245-5 2

191. Xu SS. Gao ZX, Weng 2, Du ZM, Xu WA, Yang JS, Zhang ML, Tong ZH, Fang YS. Chai XS. et al.: Efficacy of tablet huperzine-A on memory, cognition, and behavior in Alzheimer's disease. Chung Kuo Yao Li Hsueh Pao L 995; l6(5):39 1-5

192. Whitehouse PJ, Voci J: Therapeutic trials in Alzheimer's disease. Curr Opin Neurol 1995; 8(4):275-8

193. Asthana S, Raffaele KC, Berardi A, Greig NH, Haxby JV, Schapiro MB, Soncrant TT: Treatment of Alzheimer disease by continuous intravenous infusion of physostigmine. Alzheimer Dis Assoc Disord 1995; 9(4):223-32

194. Asthana S, Greig NH, Hegedus L, Holloway HH, Raffaele KC, Schapiro MB. Soncrant TT: Clinical pharmacokinetics of physostigmine in patients with Alzheimer's disease. Clin Pharmacol Ther 1995; 58(3):299-309

195. Sramek JJ, Block GA, Reines SA, Sawin SF, Barchowsky A, Cutler NR: A multiple-dose safety trial of eptastigmine in Alzheimer's disease, with pharmacodynamie observations of red blood ce11 cholinesterase. Life Sci 1995; 56(5):3 19-26

196. Minthon L, Edvinsson L, Gustafson L: Tacrine treatment modifies cerebrospinal fluid neuropeptide levels in Alzheimer's disease. Dementia 1 994; j(6):2%-3 0 1

197. Watkins PB, Zimmerman HJ, Knapp MJ, Gracon SI, Lewis KW: Hepatotoxic rffects of tacrine administration in patients with Alzheimer's disease. Jarna 1994; 27 1 ( 1 3):992-5

198. Maltby N, Broe GA. Creasey H. Jorm AF. Christensen H. Brooks WS: Efficacv o f tacrine and lecithin in rnild to moderate Alzheimer's disease: double blind. Bmj 1694: 308(6933):879-83

199. Kim YS, Nibbelink DW, Overall JE: Factor structure and reliability of the Alzheimer's Disease Assessment Scale in a multicenter trial with linopirdine. J Geriatr Psychiatry Neurol 1994; 7(3):74-83

100. Levy A, Brandeis R, Treves TA, Meshulam Y, Mawassi F, Feiler D, Wengier A. Glikfeld P. Grunwald J, Dachir S, et al.: Transdennal physostigmine in the treatment of Alzheimer's disease. Alzheimer Dis Assoc Disord 1994; 8(1): 15-2 1

20 1. Mihara M, Ohnishi A, Tomono Y, Hasegawa J. Shirnamura Y. Yamazaki K. Morishita N: Pharmacokinetics of E2020, a new compound for Alzheimer's disease. in healthy male volunteen. Int J Clin Pharmacol Ther Toxicol 1993; 3 l(5):223-9

202. Riemann D, Hohagen F, Bahro M, Lis S, Stadmuller G, Gann H, Berger M: Cholinergie neurotransmission, REM sleep and depression. .J Psychosom Res 1994; 38(Supp 1 1 ): 1 5-25

203. Sano M, Bell K, Marder K, Stricks L, Stem Y, Mayeux R: Safety and efticacy of oral physostigmine in the treatment of Alzheimer disease. Clin Neurophamiacol 1 993; 16( 1 ):6 1-9

204. Schneider LS, Olin JT, Pawluczyk S: A double-blind crossover pilot study of 1-deprenyl (selegiline) combined with cholinesterase inhibitor in Alzheimer's disease. Am .l Psychiatry 1993; 150(2):321-3

205. Holsboer-Trachsler E, Hatzinger M, Stohler R, Hemmeter U, Gray J. Muller I, Kocher R. Spiegel R: Effects of the novel acetylcholinesterase inhibitor SDZ ENA 713 on sieep in man. Neuropsychopharmacology 1993; 8(1):87-92

206. Minthon L, Gustafson L, Daifelt G, Hagberg B, Nilsson K. Risberg J, Rosen 1, Seiving B. Wendt PE: Oral tetrahydroaminoacndine treatment of Alzheimer's disease evaluated clinically and by regional cerebral blood flow and EEG. Dementia 1993; 4(1):32-42

207. Bierer LM, Aisen PS, Davidson M, Ryan TM, Stem RG, Schmeidler J, Davis KL: A pilot study of oral physostigmine plus yohirnbïne in patients with Alzheimer disease. Alzheimer Dis Assoc Disord 1993; 7(2):98-104

208. Gustafson L: Physostigmine and tetrahydroaminoacridine treatment of Alzheimer's disease. Acta Neurol Scand Suppl 1993; 149(Rand):39-41

209. Perryman KM, Fitten LJ: Delayed matching-to-sample performance dunng a double- blind trial of tacrine (THA) and lecithin in patients with Alzheimer's disease. Life Sci 1993; 53(6):479-86

210. Sahakian BJ, Coull JT: Tetrahydroarninoacridine (THA) in Alzheimer's disease: an assessrnent of attentional and mnemonic function using CANTAB. Acta Neurol Scand Supp l 1993; 14929-35

21 1. Sahakian BJ, Owen AM, Morant NJ. Eagger SA, Boddington S. Crayton L. Crockford HA, Crooks M. Hill K, Levy R: Further analysis of the cognitive effects of tetrahvdroaminoacndine (THAI in Alzheimer's disease: assessment o f attentional and rnnekonic function using CANTAB. Psychopharmacology 1993; 1 10(4):395-JO 1

212. Holford NH, Peace KE: Results and validation of a population pharmacodynamic modsl for cognitive effects in Alzheimer patients treated with tacrine. Proc Natl Acad Sci C S .-\ 1992; 89(23): 1 1471 -5

213. Holford NH, Peace KE: Methodologic aspects of a population pharmacodynamic model for cognitive effects in Alzheimer patients treated with tacrine. Proc Natl Acad Sci U S .A 1992; 89(23): L 1466-70

21 4. Cutler NR. Sramek JJ, M q h y MF, Nash RJ: Implications of the study population in the early evaluation of anticholinesterase inhibitors for Alzheimer's disease. Ann Phmacothrr 1992; 26(9): 1 1 1 8-22

21 5 . Cutler NR, Sramek JJ, Murphy MF, Nash RJ: Alzheimer's patients should be includrd in phase 1 clinical trials to evaluate compounds for Alzheimer's disease. J Geriatr Psychiatry Xeurol 1992; 5(4): 192-4

216. Cohen MB, Fitten LJ, Lake RR, Perryman KM, Graham LS. Sevrin R: SPECT brain imaging in Alzheimer's disease dunng treatment with oral tetrahydroarninoacridine and lecithin. Clin Nucl Med 1992; 1 7(4):3 L 2-5

217. Eagger S, Morant N, Levy R, Sahakian B: Tacnne in Alzheimer's disease. Time course of changes in cognitive function and practice effects. Br J Psychiatry 1992; 160:36-40

218. Eagger SA, Levy R, Sahakian BJ: Tacrine in Alzheimer's disease. Acta Neurol Scand Suppl 1992; 139:75-80

219. Eagger SA, Levy R, Sahakian BJ: Tacrine in Alzheimer's disease [see comments]. Lancet 199 1 ; 337(8748):989-92

220. Ebmeier W, Hunter R, Curran SM, Dougal NJ, Murray CL. Wyper DJ. Patterson J. Hanson MT, Siegfried K, Goodwin GM: Effects of a single dose of the acetylcholinesterase inhibitor veinacnne on recognition memory and regional cerebral blood flow in Alzheimer's disease. Psychopharmacology 1992; 108(1-2): 103-9

221. Perryman KM, Fitten LI: Quantitative EEG during a double-blind trial of THA and lecithin in patients with Alzheimer's disease. J Genatr Psychiatry Neurol 199 1; 4(3): 127-33

222. Sunderland T, Molchan S, Lawlor B, Martinez Et, Mellow A, Martinson H, Putnarn K, Lalonde F: A strategy of "combination chernotherapy" in Alzheimer's disease: rationale and

preliminary results with physostigmine plus deprenyl. Int Psychogeriatr 1992: ?(Suppl 2 ) : D 1-309

223. T a c ~ e as a treatment for Alzheimer's disease: editor's note. An interim-report from the FDA. A response fiom Summers et al. N Engl J Med 1991; 324(5):349-52

224. Huff FJ. Antuono P, Murphy M, Beyer J, Dobson C: Potential clinical use of an adrenergic/cholinergic agent (HP 128) in the treatment of Alzheimer's disease. .ein N Y Xcad Sci 199 1; 640243-7

225. Molloy DW, Guyan GH, Wilson DB, Duke R, Rees L. Singer J: Effcct of ietrahydroaminoacridine on cognition, function and behaviour in Alzheimer's disease. Cmaj 1991; 144(1):29-34

126. Schneider LS, Lyness SA. Pawluczyk S, Gleason RP, Sloane RB: Do blood pressure and age predict response to tacrine (THA) in Alzheimer's disease? A preliminary report. Psychopharmacol Bull 1991; 27(3):309- 14

227. Tune L, Brandt J. Frost JJ, Hams G, Mayberg H, Steele C, Burns A. Sapp J. Folstein MF. Wagner HN, et al.: Physostigmine in Alzheimer's disease: effects on cognitive functioning. cerebral glucose metabolisrn analyzed by positron ernission tomography and cerebral blood tlow analyzed by single photon emission tomogaphy. Acta Psychiatr Scand Suppl 199 1 : 366:O 1-5

228. Puri SK, Ho I, Hsu R, Lassman HB: Multiple dose pharmacokinetics. safety. and tolerance of velnacrine (HP 029) in healthy elderly subjects: a potential therapeutic agent for Alzheimer's disease, J Clin Phmacol 1990; 3O(l O):948-55

229. Puri SK, Hsu RS, Ho I, Lassman HB: The effect of food on the bioavailability of velnacrine (HP 029) in healthy elderly men: a potential Alzheimer agent. J Clin Pharmacol 1989; 29( 10):956-60

230. Puri SK, Hsu RS, Ho 1, Lassman HB: Single dose safety, tolerance, and pharmacokinetics of HP 029 in healthy young men: a potential Alzheimer agent. I Clin Pharmacol 1989: 19(3):278-84

231. Cutler Mi, Murphy MF, Nash RJ, Prior PL, De Luna DM: Clinical safety, tolerance. and plasma levels of the oral anticholinesterase 1,2,3,4-tetrahydro-9-arninoacridin- 1 -oL-maleate (HP 029) in Alzheimer's disease: preliminary findings. .J Clin Pharmacol 1990; 3O(6):5 56-6 1

232. Gauthier S, Bouchard R, Lamontagne A, Bailey P, Bergman H, Ratner I, Tesfaye Y. Saint-Martin M, Bacher Y, Camer L, et al.: Tetrahydroaminoacridine-lecithin combination treatment in patients with intermediate-stage Alzheimer's disease. Results of a Canadian double- blind, crossover, multicenter study. N Engl J Med 1990; 322( 18): 1272-6

233. Chatellier G, Lacomblez L: Tacrine (tetrahydroarninoacndine; THA) and lecithin in senile dementia of the Alzheimer type: a multicentre trial. Groupe Francais d'Etude de la Tetrahydrûarninoacridine. Bmj 1990; 300(6723):495-9

234. Davies B, Andrewes D, Stargatt R, Ames D, Tuckwetl V, Davis S: Tacrine in Alzheimer's disease [letter]. Lancet 1989; 2(8655): 163-4

735. Jenike MA, Albert M, Baer L, Gunther J: Oral physostigmine as treatment for primary degenerative dementia: a double-blind placebo-controlled inpatient trial. J Grriatr Psychiatry Neurol 1990; 3(l): 13-6

236. Molloy DW, Cape RD: Acute effects of oral pyridostigmine on memory and cognitive function in SDAT. Neurobiol Aging 1989; 1 O(2): 199-204

237. Mohs RC, Davis KL: A signal detectability analysis of the effect of physostigmine on memory in patients with Alzheimer's disease. Neurobiol Aging 1982; 3(2): 105- 10

238. Chnstie JE, Shering A, Ferguson J, Glen Al: Physostigmine and arecoline: effects of intravenous infusions in Alzheimer presenile dementia. Br J Psychiatry 198 1 ; 1 38:46-50

239. Stem Y, Sano MT Mayeux R: Long-term administration of oral physostigmine in Alzheimer's disease. Neurology 1 988; 38( 12): 1 83 7-4 1

240. Surnmers WK, Majovski LV, Marsh GM, Tachiki K. Kling A: Oral tetrahydroaminoacridine in long-term treatment of senile dementia, Alzheitner type. N E n j l J bled 1986; 3 ls'(2O): 1241-5

241. Smith CM, Swash M: Physostigmine in Alzheimer's disease [letter]. Lancet 1979: l(8 1 O6):4S

232. Thal LJ, Masur DM, Sharpless NS, Fuld PA, Davies P: Acute and chronic rtTects of orül physostigmine and lecithin in Alzheimer's disease. Prog Neuropsychop harmaco l B io l Psychiatry 1986; 1 O(3-5):627-36

243. Murphy MF, Hardiman ST, Nash RI, Huff FJ, Demkovich IJ, Dobson C, Knappr CE: Evaluation of HP 029 (velnacnne maleate) in Alzheimer's disease. Ann N Y Acad Sci 199 1: 040253-62

2 . Wilcock GK, Surmon DJ, Scott MT Boyle M, Mulligan K. Neubauer KA. O'Neill D. Royston VH: An evaluation of the efficacy and safety of tetrahydroarninoacndinr (THA) without lecithin in the treatment of Alzheimer's disease. Age Ageing 1993; 23(5):3 16-14

215. Jelic V. Dierks T, Arnberla K, Almkvist O, Winblad B, Nordberg A: Longitudinal changes in quantitative EEG dunng long-term tacrine treatment of patients with Alzheimer's disease. Neurosci Lett 1998; 254(2):85-8

246. Kumar V, Anand R, Messina J, Hartman R Veach J: An efficacy and safety analysis of Exelon in Alzheimer's disease patients with concurrent vascular risk factors. Eur J Yeurol 2000; 7(2): 159-69

147. Xu SS, Cai ZY, Qu ZW, Yang RM, Cai YL, Wang GQ, Su XQ, Zhong XS, Cheng RY. Xu WA, Li IX, Feng B: Hupenine-A in capsules and tablets for treating patients with Alzheimer disease Cpublished erratum appears in Chmg Kuo Yao Li Hsueh Pao 1999 Aug;20(8):732]. Chung Kuo Yao Li Hsueh Pao 1999; 20(6):486-90

248. UNU L, Vicari S, Moriearty P, Schaefer F, Becker R: The recovery of cerebrospinal fluid acetylcholinesterase activity in Alzheimer's disease patients after treatment with metrifonate.

249. Shikiar R, Shakespeare A, Sagnier PP, Wilkinson D, McKeith 1, Dartigues K. Dubois B: The impact of metri fonate therapy on caregivers of patients with Alzheimer's disease: results fkorn the MALT clinical trial. Metrifonate in Alzheimer's Disease Trial. J Am Geriatr Soc 2000; 48(3):268-74

250. Gelinas I, Gauthier S, Cyrus PA: Metrifonate enhances the ability of Alzheimer's disease patients to initiate, organize, and execute instrumental and basic activities of daily living. J Geriatr Psychiatry Neurol2000; l3(l):9- 1 6

251. Blass JP, Cynis PA, Bieber F, Gulanski B: Randomized. double-blind. placebo- controlled, muhicenter study to evaluate the safety and tolerability cf mtrifun~titr in patients with probable Alzheimer disease. The Metri fonate Study Group. Alzheimer Dis Assoc Disord 2000; 14(1):3945

252. Kaufer D: Beyond the cholinergie hypothesis: the effect of metnfonate and othrr cholinesterase inhibitors on neuropsychiatx-ic symptoms in Alzheimer's disease. Dement Geriatr Cogn Disord 1998; 9(Suppl2):8- 14

253. Weinstein HC, Ternisse S, van Go01 WA: Tetrahydroarninoacridine and lecithin in the treatment of Alzheimer's disease. Effect on cognition, functioning in daily life. behavioiiral disturbances and burden experienced by the caren. J Neurol 1 99 1 ; 238( 1 ):34-8

251. h t u o n o PG: Effectiveness and safety of velnacrine for the treatment of Alzheimer's disease. A double-blind. placebo-controlled study. Mentane Study Group. Arch h e m Med 1995; 155(16): 1766-72

3 5 . Becker RE. Colliver JA, Markwell SJ. Morîearty PL, Unni LK. Vicari S: Double-blind. placebo-controlled study of metrifonate, an acety lcholinesterase inhibi tor. for Alzheimer disease. Alzheimer Dis Assoc Disord 1996; 1 O(3): 124-3 1

256. Becker RE, Colliver JA, Markwell SJ, Moriearty PL, Unni LK, Vicari S: Effects of metri fonate on cognitive decline in Alzheimer disease: a double-blind, placebo-contro lled, 6- month study. Alzheimer Dis Assoc Disord 1998; 12(1):54-7

257. Cummings .IL, Cyrus PA, Bieber F, Mas J, Orazem J, Gulanski B: Metrifonate treatment of the cognitive deficits of Alzheimer's disease. Metrifonate Study Group. Neurology 1998; jO(5): 12 1 4-2 1

258. Moms JC, Cyrus PA, Orazem J, Mas J, Bieber F, Ruzicka BB, Gulanski B: Metrifonate benefits cognitive, behavioral, and global hinction in patients with Alzheimer's disease [see comments]. Neurology 1998; SO(5): 1222-30

259. Forette F, Anand R, Gharabawi G: A phase II study in patients with Alzheimer's disease to assess the preliminary eficacy and maximum tolerated dose of rivastigmine (Exelon). Eur J Neurol 1999; 6(4):423-9

260. Imbimbo BP, Martelli P, Troetel WM, Lucchelli F, Lucca U, Thal LJ: Efficacy and safety of eptastigmine for the treatment of patients with Alzheimer's disease. Neurology 1999; 52(4):700-8

26 1. hb imbo BP, Lucca U, Lucchelli F, Alberoni M, Thal LJ: A 25-week placebo-controlled study of eptastigmine in patients with Alzheimer disease. Alzheimer Dis Assoc Disord 1998; 1 ?(4):3 13-22

262. Thal LJ, Ferguson JM, M i n ~ e r J, Raskin A, Targum SD: A 24-week randomized trial of controlled-release physostigmine in patients with Alzheimer's disease. Neurology 1999; 52(6): 1 146-52

263. van Dyck CH, Newhouse P, Falk WE, Mattes JA: Extended-release physostigmine in Alzheimer disease: a multicenter, double-blind, 12-week study with dose enrichment. Physostigmine Study Group .kch Gen Psychiatry 2000; 57(2): ! 57-64

264. Wood P. Castleden M: k double-blind, placebo, controlled, multicentre study of tacrine For Alzheimer's disease. International Journal of Geriatric Psychiatry 1994; 9549-54.

265. Corey-Bloom J, Anand R, Veach J. Group. tEBS: A randomized trial svaluating the eficacy and saftey of' ENA 713 (rivastigrnine tartrate), a new acetylcholinesterase inhibitor. in patients with rnild to moderately severe Alzheimer's disease. International Journal of Geriatric Psychopharmacology 1998; 155-65.

266. Agid Y, Dubois B, Investigaton. tiR, Anand R, Gharabawi G: Eficacy and tolerability of nvastigmine in patients with dementia of the Alzheimer type. Current Therapeutic Research 1998; 59( l2):837-845.

267. Schneider LS, Farlow MR, Henderson VW. Pogoda SM: Effects of estrogen replacement therapy on response to tacnne in patients with Alzheimer's disease. Neurology 1996: 46(6): 1 580-4

168. Schneider LS, Farlow M: Combined tacrine and estrogen replacement therapy in patients with Alzheimer's disease. Ann N Y Acad Sci 1997; 826:3 17-22

269. Clipp EC, Moore MJ: Caregiver time use: an outcome measure in clinical trial research on Alzheimer's disease. Clin Pharmacol Ther 1995; 58(2):328-36

270. Imbimbo BP, Nicoli M, Martini C, Tomellen GP, Martelli P, Ferrari GP. Garnbina G: Acetylcholinesterase assay may predict cognitive response of Alzheimer patients to eptastigmine treatment [letter]. Eu . J Clin Pharmacol 1998; 54(9- LO):809- 10

271. Allain H, Schuck S, Lebreton S, Strenge-Hesse A, Braun W. Gandon JM. Brissot P: Aminotransferase Levels and silymarin in de novo tacrine-treated patients with Alzheimer's disease. Dement Geriatr Cogn Disord 1999; 1 O(3): 18 1-5

Appendix A - Consent form for Study 1.1

Suanvbroak and W n e b ' $ Callqc Hcalth 5cienctc Centre

PSYCHOPHARMACOLOCY

RESEARCH PROGRAM 2075 Bayview Avc Rrn. F3 27 Toronto. Ontario M4N 3M5

Dr. CIaudio A. Naranjo. MD Head, Psychopharmacolog)l

Researcti Program

Dr. Nathan Hemnmn. :MD, FRCPC

Head, Dtvision of Gerratrrc Pqcftrarry, S& WCHSC und Lin1 verstry of Toronto

Kcnncth Yau. BSC .Wasttv-'s Srudenr.

C~'nrversi~ of Toronfo. Psi.ctropirurn~acolog):

Researcil Program. S& WCHSC

Tel.: (41 6 ) 580-6 100 x.3 185 Fa\.: (4 16) 480-4708 E-mai 1: ken. yau@swc hsc.on.ca

-

U n i v e r s i t y

o f T o r o n t o

Developrnent of a Peripheral Marker to Assess the Sensitivity of the Central Nervous System to Cholinergic Medications:

PILOCARPINE TEST ONLY

Investigators: C.A. Naranjo, MD, N. Herrmann MD and K. Lanctôt PhD

Study Co-ordinator: Kenneth Yau (416) 480-6100 Extension 3 185

SUBJECT INFORMATION/CONSENT FORM

DESCRIPTION AND PURPOSE OF THE STUDY You have been asked to participate in a research study involving the use of a medication called pilocarpine. The purpose of this study is to mesure the effects of pilocarpine on your pupil size.

PROCEDURES This study will require that you corne to Sunnybrook for one half day (2 hours) study session.

Study VLFirs: You will be asked to corne to roorn FG53a at Sunnybrook. Your eyes will be tested in a similar manner to an rye test at the optometrist's office. Eye Tests: You will be given an eye drop of pilocarpine in one eye. and a drop of placebo (saline) in the other eye. All eye drops will be administered with a precision dropper. Measurements of both eyes will be made with a video camera that takes a picture of your pupil. These measurements will be made four times within a 1 hour prriod. Nothing touches your eye when this is done. E ~ r a Procedures: You will be asked to have an eye test for glaucoma.

PARTICIPATION AND TERMINATION Your participation in this trial is voluntary. You may withdraw From the study at any time. Your rehsal to participate will have no influence on your usual medicd care.

RISKS Al1 therapies have the potential to cause some adverse experiences. Pilocarpine is a commonly used dnig for pupil constriction and treatment of glaucoma. You rnay expenence mild imtation to the eye such as stinging or itching, fiom the eye &op. Other side effects which are fiequent (10-70%) may dso occur: temporas, pupil contraction, fuzzy vision, and near sightedness. Although the dmg is administered only to the eye, there may be other parts of your body

Page li4

that rnay be affected and you rnay notice effects like increased sweating. These side effects occur less frequently (IO%), are temporary and will be minimized by using doses of the eye drops that are lower thm the doses employed in regular health care settings. Should you. as a result of the study dmg or the research procedures, be h m e d in any way. you will receive appropriate emergency or other medical treatment.

CONFIDENTIALITY The medical information gathered in this study will be kept confidential. Only medical professionals will have access to your records, and should these results be published you will not be identified in any way.

QUESTIONS REGARDING PARTICIPATION If you have medical or study related questions, or feel you have had a research related injury. !ou may contact the Office of Research Administration at (416) 480-4293.

SUBJECT'S iNFORMED CONSENT PROTOCOL 5077rW0039

1, was given a copy of this signed and dated Informed Consent F o m and the corresponding Subject Information Sheet. I have received an explmation of the nature, purpose, duration and foreseeable effects of the trial and what I will be expected to do. The possible nsks and benefits of the trial have been explained to me. I was +en time and opportunity to consider the above information and a11 my questions were 3

answered to my satisfaction.

1 vùluniarilÿ consent to participate in tliis triai.

Subject 's Signature Date (dd/rnm/yy)

1 confinn that 1 have explained the nature. purpose and foreseeable effects of the mat to the subjecr whose name is pnnted above. The subject consented to participate by hisiher psrsonally dated signature.

In t'onned Consent Provrder's Signature Date (dd/mm/yy)

Informed Consent Provider's Name (pnnted)

Page 31'4

STATEMENT OF WITNESS TO LNFORMED CONSENT OF SUaTECT

I confirm that 1 have explained the nature, purpose and foreseeable effects of the trial to the subject whose narne is printed above. and that helshe has voluntady consented to participate. 1 agree to co- operate fully with the supervising doctor and will tell him if the subject suffen any unexpected or unusual symptoms. I confirm that 1 have informed the supervising doctor of any medication/dnig, of whatrver nature. the subject has taken in the month preceding the stan of the trial. or is taking or plans to take. whether prescribed or not. i understand that the subject is free to withdraw from the nial a< any tirne. without the need to justify this decision and without any disadvantage to the further medical treatment. 1. have witnessed the expianation of the information contamed in the Subject Information Sheet, to the following subject: - . This subiect l.oluntariIy agreed to take part in the trial.

Subjec t Representative's Signature Date (dd/mm/yy)

Subj ec t Representative's Name (pnnted) Relationship to Subject (printsd)

1 continn that I have explained the nature. purpose and foreseeable cffects of the mal to the subject whose narne is printed above. The subject consented to participate by hisher personally dated signature.

informed Consent Provider's Signature

[nformed Consent Provider's Name (printed)

Page 4 4

Appendiw B - Consent form for Study 1.2

Sunnvbrwk and Hliorncn's Collqe Hcilth ktences Ccntn

PSYCHOPHARMACOLOGY RESEARCH PROGRAM 1075 Bayview Ave Rm. F3 27 Toronto, Ontario M W 3M.5

Dr. Claudio A. N m j o , M D Httarf, Psyclmpharntacology

Reseurcii Prograrn

Dr. Nathan Hemnann. MD, FRCPC

Head, Division of Geriafric Psychratry, S& WCHSC and Lbriversil of Toronto

Kcnnrth Yau. BSc Muser 's Scudent.

LLniversity of Toronto. Psvchopir urmacolog~.

Research Prograrn. Sc3 WCHSC

Tel.: (4 16) 480-6 1 O0 x.3 185 Fm.. (4 16) 480-4708 E-mail: ken.yau@swchsc.on.ca

DONEPEZIL IN ALZHEIMER'S STUDY: PIL-CT TEST

Investigaton: C.A. Naranjo MD, N. Hermann MD, and K. Lanctôt PhD

Shidy Co-ordinator: Kenneth Yau (416) 480-6100 Extension 3185

SUBJECT INFORR24TIONlCONSENT FORkI

DESCRIPTION AND PURPOSE OF THE STUDY You have been asked to participate in a research study involving the use of two medications: donepezil and pilocarpine. The purposr of this study is to measure your pupil size to see if changes in your eye can be used to measure response to dmg treatments for the memory losses you have been experiencing called Alzheimer's disease. This study will also determine the type of one of your genes. a gene which rnakes a protein in your body called ApoE. The purpose of determining what type of gene you have (this process is callrd genotyping) is that this gene c m affect the hnction of the pupil of your eye which occurs foilowing the drugs given in this study.

PROCEDURES This study will require that you corne to Sunnybrook for 3 half-day (2.5-hour) study sessions. You will receive medication (donepezil) for Alzheimer's disease. Study Treatments: You will receive 4 weeks of treatment with 5 - r n ~ donepeziVday foilowed by 8 weeks of treatment-wi th 1 O-mg donepezil/day. Study Visits: You will be asked to corne to mom EG 04 at Sunnybrook. Your eyes will be tested in a similar rnanner to an eye test at the optometrist's office. In addition, you will be given a memory test, and a blood sample will be taken. Eye Tests: You will be given an eye drop of pilocarpine in one eye. and a drop of placebo (saline) in the other eye. Al1 eye drops will be administered with a precision dropper. Measurements of both eyes will be made with a video camera that takes a picture of your pupil. These measurements will be made four times within a 1-hour penod. Nothing touches your eye when this is done. Memory Tests: In addition, you will be asked to do a simple rnemory test, in which you will be asked to remember some simple words and phrases, as well as some pencil and paper tests. in addition. you (or your caregiver) will be asked some questions about your activities of daily Living.

U n i v e r s i t y

o f T o r o n t o

Page 1!5

Blood Samples: A blood sample (2 tubes of blood, about two tablespoons) will be taken. The purpose of this blood sample will be to perform routine analysis of your blood chemistry and haematology. in addition, a blood sample will be analyzed to see if you have a gene which has been associated with changes in pupil size (ApoE alleles) - see below. Genotyping: An additional blood sample (one tube, about one tablespoonful) will be taken to see if you have a certain gene (apolipoprotein E ~ 4 ) that may cause an altered pupil response in your eyes dunng the eye tests. This process is called genotyping, and we will analyze the blood sample to see which of the three types of the gene coding for a certain protein callsd apolipoprotein E (ApoE) you have. Since researchers have shown that your ApoE gene type affects pupil responses when eye drops are applied to your eyes. we are only interested in studying whether different types of this gene may predict how your eyes will respond to the pilocarpine eye drops. Recent studies suggest that different combinations or types of this gene may be used to predici the risk of developing Alzheimer's disease. however, determining your genotype is not an established genetic test for Alzheimer's disease. In other words, researchers today cannot tell you or your farnily anything about Alzheimer's disease based on the variant of the ApoE gene you have. Results of Genotyping: Your blood samples will be stored until the genotyping is perfomed for this study. AAer that, they will be destroyed. Your blood samples will not be used for any othcr purpose. You rnay decide to withdraw your samples from the study at any time. In the event of your death, your sample will only be used to obtain the information that you gave consent to. and the sample will be destroyed f i e r the test. Your blood sample will be stored by the Psychopharmacology Research Program until genotyping is performed.

Exira Procedures (First Visit): You will be asked to have an eye test for glaucoma, and a test of your heart f i c t ion (an electrocardiogram: ECG). Additional Information: During the trial you will be asked to take the study medication once daily in the evening. For this trial to be successful, it is important that you CO-operate fully with your doctor and follow hisher instructions precisely. You may be asked to stop taking some of your current medications. Also, you will be asked not to take any other dmgs or remrdies rither on prescription, or from the pharmacy or supermarket unless it has been approved by the doctor conducting the trial.

PARTICIPATION iUVD TERMINATION Your participation in this trial is voluntary. You may withdraw from the study at any time. Your refusal to participate or wish to withdraw before completion of this trial penod will have no influence on your usual medical care. Your participation in this study cm be terminated without your consent by the study doctor. if it is deterrnined to be in your best interest or if you fail to follow the directions of the doctor. During each visit you will be asked about any adverse events you may have experienced during the trial since the last visit.

RISKS By participating in this study, the evolution of your condition will be closely monitored with more care than usual. Risks associated with participation in this study include potential for rare instances of infection or bniising and occasional fainting nom bIood drawing. Al1 therapies have the potential to cause some adverse experiences.

Page 2 5

The most cornmon side effects observed fiom studies with donepezil occurred in up to 10?/a of patients, or up to 1 in 10 patients experienced the following: nausea (6%), diarrhea (99'0). insomnia (6%). or stomach problems (5%). Previous studies with this dnig showed that patients generally experienced mild symptorns over a short period of tirne that cleared up while the patient continued to take donepezil. Other possible side effects include muscle crarnps (3%). fatigue (3%), and weight loss (3%). Lastly, the risk of donepezil to the hurnan embryo or fetus has not been identified. Pilocarpine is a comrnonly used dmg for pupil constriction and treatment of glaucoma. You rnay experience mild irritation to the eye such as stinging or itching, kom the eye drop. Othcr side effects which are frequent (10-70%) may also occur: temporary pupil contraction. fuzzy vision. and near sightedness. Although the dmg is administered only to the eye, there may be other parts of your body that may be affected and you may notice effects like increased sweating. These side effects occur less Eequently (IO%), are temporary and will be rninimized by using doses of the eye drops that are lower than the doses employed in regular health care settings. Should you, as a result of the study dmg or the research procedures, be harmed in any way. you will receive appropriate emergency or other medical treatment. During the trial, you will be told if additional information is discovered that rnay affect your willingness to continue in this trial.

ALTERNATIVE TREATMENTS Although donepezil is currently the only medication of its type available for Alzheimer's disease. alternative treatments for your illness are available. These can be discussed with your doctor. You have the nght to ask any questions concerning the potential and/or known hazards of this trial at any time.

CONFIDENTIALITY The rnedical information gathered in this study will be kept confidential. Only medical prolessionals will have access to your records, and should these results be published you will not be identified in any way. Coafidentiality of Genetic test results: The blood samples for ApoE genotyping will br labelled with your initials and date of birth. The label will not contain your name. You will be identifiable as an individual by the principal investigator: Dr. C. A. Naranjo. and his co- investigators in this study. The results from your genotyping will not be made available to any person, family member or institution without your authorkation in writing. You may provide us with written consent to disclose your genotyping results to your family physician who may discuss the results with you. Counselling for the interpretation of your genotyping results will be available fiorn Dr. N. Hemnann, who is a genatric psychiatrîst and CO-investigator in this study. Your blood sample and genetic results will not be shared with any other institutions or individuals unless written consent was provided by you. You will be contacted for your consent in the unlikely event that the samples are io be used in any way that you had not been made aware, or that you had not given consent.

QUESTIONS REGARDING PARTICIPATION If you have medical or study related questions, or feel you have had a research related injury, you may contact the Office of Research Administration at (4 16) 480-4293,

Page 315

SUBJECT'S INFORMED CONSENT PROTOCOL s07n~10039

1. was given a copy of rhis signed and dated Informed Consent Form and the corresponding Subject Information Sheet. 1 have received an explanation of the nature, purpose, duration and foreseeable effects of the trial and what 1 will be expected to do. The possible nsks and benefits of the trial have been explained to me. 1 was given time and opportunity to consider the above information and al1 my questions were answered to my satisfaction.

1 voluntarily consent to participate in this trial.

Subject's Signature Date (dd/mm&y)

1 confirm that 1 have explained the nature, purpose and foreseeable effects of the trial to the subject whose name is printed above. The subject consented to participate by hisiher penonally dated signature.

Infomed Consent Provider's Signature Date (dd/mmyy)

Infomed Consent Provider's Name (printed)

STATEMENT OF WITNIESS TO INFORMED CONSENT OF SUBJECT

1 confirm that 1 have explained the nature, purpose and foreseeable effects of the trial to the subject whose name is printed above, and that he/she has voluntarily consented to participate. t agree to CO-operate hlly with the supervising doctor and will tell him if the subject suffers an? unexpected or unusual symptoms. 1 confirm that 1 have informed the supervising doctor O f any medicatioddnig, of whatever nature, the subject has taken in the month preceding the start of the trial. or is taking or plans to take, whether prescribed or not. 1 understand that the subject is Free to withdraw barn the trial at any time, without the need to justify this decision and without any disadvantage to the furthet medical treatmen t . 1. have witnessed the explanation of the information containrd in the Subject Information Sheet, to the following subject: This subject voluntarily agreed to take part in the trial.

Subject Representative's Signature Date (ddimm/yy)

Subject Representative's Name (printed) Relationship to Subject (prhted)

I confirm that I have explained the nature, purpose and foreseeable èffects of the trial to the subject whose name is pnnted above. The subject consented to participate by his/her personally dated signature.

Inforrned Consent Provider's Signature Date (dd/mrn/yy )

- -

Infomed Consent Provider's Name (printed)

Page si5

Appendix C - Inclusion form for meta-analysis

REVIEWER'S INITIALS 113.1

PAPER# KY-

INCLUSION CRITERIA FOR ORAL CHOLINESTERASE INHIBITORS IN ALZHEIMER'S DISEASE

Based on Methods section only

1. C h i c d tiial: Was the study an original clinical trial?

2. Randomized: Was there randorn docation to treatment groups?

3. Controlled: Was this study controlled by either placebo or a cornparison medication?

4. DoubIe blind: Did this study use double blind techniques?

5. Parallel groups: Did this study use a parallel group design?

- -

6. Diagnosis of Alzheimer's disease: Were patients assessed for the presence of AD? (e.g., DSM-IV a d o r NNCDS-ARDRA)

7. Medications: Were patients treated with an oral cholinesterase inhibitor?

8. Length of double blind treatment: Were patients treated for at least 12 weeks with the medication in a double blind design?

9. Cognitive outcornes: Was the eficacy of the medication assessed for cognitive improvement? (any standardized scale acceptable)

If the paper meets al1 inclusion cntena it is to be accepted. If accepted, please read Results sections and fil1 out quality assessrnent and data extraction forms.

AppendLu D - Quality Assessment form for meta-analysis

QUALITY ASSESSMENT FOR ORAL CHOLINESTERASE INHIBITORS IN ALZHEIMER'S DISEASE

Quality Legend

O = not reported 1 = poor 2 = satisfactory 3 = good

1 Stirdy Subjects I i l

1 . Description of study setting and source ofsubjects (e.g., type of institution, level 1 of eue, referral pattern, length of thne on unit)

I 2. Appropriateness of eligibility cntena (explicit inclusion & exclusion criteria, clarity )

] 3. Suitability of the cornparison group (were the groups alike?) 1 1

5. Blinding of çubjects (to what degree was it achieved?) i

6. Blinding of investigators (to what degree was it achieved?)

1 7. Description of study procedures 1 1 8. Control of concomitant medications (were important psychoactive dmgs 1 pmhibited and concomitant medications taken recordai?)

- - - -- - - - - - -

9. Definition of cognitive outcome criteria (operationalized? validated?) j I 1

Results 1 7

4

10. Information on eligible subjects not enrolled (reasons & numbers) I ! 1 1 1. Adequacy of description of enroiled sample (comorbidities & diagnosis of AD) 1 1

12. Description of &op-outs I 1

13. Description of adverse events -

An alysis 1 14. SampIe size justification

)15.0nitoring of cornpliance with objective measures 1 1 ] 16. Appropnateness of statistical analysis of results 1

--

Overail qualig

17. Overall qyality of papa (global rating independent of previous scores)

AppendLu E - Data Extraction form for meta-analysis

REVIEWER'S INITIALS

DATA EXTRACTION FOR ORAL CHOLINESTERASE INHIBITORS IN ALZHEIMER'S DISEASE

1. GENERAL INFORMATION

Which cognitive scale(s) wadwere used?

ADAS-cog

MMSE

Other (please speciQ)

What was the duration of the double blind portion of study? (# of weeks)

Treatment

Subject Age (years, mean + SD/SEM) k

Subject Age Range (years) -

Duration of lllness (yrs, mean + SDISEM) t

Range of Duration ( y e m ) - - -

Seventy of Illness* (mean + SD/SEM) t

*Spectfi (e.g. meart score on at entry)

A) Clroiin esterase l n hibitor

Baseline ADAS-cog*

Final ADAS-cog* Score (mean f SD) - + k 2

Baseline MMSE* Score (mean t SD) k t k

Final MMSE* Score (mean + SD) k t - +

*Speczfi and use other cognitive scczle(s) ifADAS-cog a n h r MiZ.ISE not tised

B) Piacebo ADAS-cog* MMSE*

Baseline Score (mean f SD)

Final Score (mean f SD)

Page 1 of 3

2. EFFICACY RATE

Indicate the studyltrial completors who responded to the treatrnent and defi ne improvemenr (e.;. +point improvement in ADAS-cog fiom baseline).

De finition of Improvement:

3. SAFETY RATE

Indicate the nurnber of studyltrial completors who completed the study who reported adverse events (AE). Please specify type of AE reported (e.g. serious vs. all)

Page 2 of 3

Treatment Group

1

I

To ta1

Number reporting AE

Number not reporting AE Total treated

4. DROP-OUT RATE

Indicate the number of subjects who enrolled in the study (Le. intention to treat) but later dropped out of the trial.

5. DROP-OUT RATE DUE TO ADVERSE EVENTS

tndicate the number of subjects who enrolled in the study (Le. intention to treat) but dropped out of the trial due to adverse events.

Page 3 of 3

Treatment Group

Total

Number not Dropped out

Number of AE Dropouts Total treated

Appendix F - Meta-analysis spreadsheet for comparative calculations

Comparative Emcacy Calculations

Cognitive Efficacy - ADAS-cog

Nt N-EVENTS NX-EVENTS NC Rt R C % E V E N T %NOT Paper ChEl+ ChEl- lacebo Placebo (A+B) (A+C) (B+D) (C+D) Subject A/(A+B C/(C+D) A(A+C) C/(A+C)

KY-1 137 165 41 112 302 178 277 153 455 0.4536 0.268 0.7696629 0.2303

KY-IO 80 223 12 74 303 92 297 86 389 0.264 0.1 3'15 0.8695652 0.1304

KY-11 93 391 39 199 484 132 590 238 722 O. 1921 0.1639 0.7045455 0.2955

KY-12 67 150 9 47 217 76 197 56 273 0.3088 O. 1607 0.8815789 0.1 184

KY-25 64 174 36 145 238 1 O0 3 19 181 419 0.2689 0.1989 0.64 0.36

KY-29 83 165 26 131 248 1 09 296 157 405 0.3347 0.1656 0.7614679 0.2385

KY-30 152 267 44 181 419 196 448 225 644 0.3628 0.1956 0,7755102 0.2245

Total = 676 1535 207 889 221 1 883 2424 1096 3307 2.1849 1.2921 5.4023306 1.5977

Avg = 96.57 21 9.3 29.57 127 3 16 126. MZ86 346.285714 157 472.4 0.3 12 1 0.1846 0.77 1761 5 0.2282

STD = 3 1.88 79.75 13.17 50.53 92.4 41.85 1535 12 1.026224 62.1 144.5 0.0772 0.039 0.0789475 0.0789

ORi SINGLE GROUP AD/B var Tx var C var DIFF si 2 wi wiRi wi(Ri-r)2 wi2 wi* wi*Ri

R* = 0.292623 var R* = 0.007645

ChEI META AVG = 0.3 1 1 142 SE META = 0.034558

LL = 0.243408 UL = 0.378876

SECOND GROUP DIFF TX si2 wi wiRi wi(Ri-r)2 wi2 wi* wi*Ri k RI - Rc wi DIFF wi*Ri

R* = 0.18421 1 var R* = 0.000436

Placebo META AVG = 0.184829 SE META = 0.01433

L1. = 0. 156743 U1, = 0.313016

HOMOG wi(R-R*)2 wi2 w*i w*iRi

META DIFFERENCE = 0.123487 SE DIFF = 0.025593

LL = 0.073324 UL = 0.173649

Global Assessrnent Efficacy - CIBIS+/CGIC

Ni N-EVENTS NX-EVENTS NC Rt R C %EVENT 'i/oNOT Paper ChEl+ ChEl- lacebo Placebo (A+-B) (A+C) (B+D) (C+D) Subjjeci A/(A+B C/(C.-D) A(A+C) C/(A+C)

KY-1 KY-7 KY-1 1 KY-13 KY-14 KY -2 1 KY-24 KY-25 KY-28 KY-29

Total =

Avg =

STD = 33.78 100.5 8.933 48.68 136 41.Z6327 147.154205 54.5 177.6 0.08 16 0.05 1 0.0587265 0.0587

ORi SINGLE GROUP ADIB var Tx var C var DIFF si2 wi wiRi wi(Ri-r)2 w i2 wi* wi*Ri

2309.856 9 1 8.9258 2045.438 1339.2 16 1759.305 1804.457 239.3695 1339.42 1 3073.779 2426.37

17256.14 1 725.61 4 769.266 1

R* =

var R* =

META AVG = 0.271 157 SE META = 0.02 1733

LL = 0.22856 UL = 0.3 13754

SECOND CROUP DIFF TX si2 wi wiRi wi(Ri-r)2 w i3 wi* wi*Ri k R t - R c wi DlFF wi*Ri

R * = 0.161201 var R* = 0.001 O42 SE META = 0.01 3805

LL - O. 140694 UL = O.1948OO

R* = 0.092452

var R* = 0.00027

HOMOG wi(R-R*)2 wi2 w*i w*iRi

META DIFFERENCE = 0.093237 SE DlFF = 0.012951

LL = 0.067854 UL - 0.1 18621

Comparative Safety Analyses

Adverse Events Nt N-Events NX-Events Nc

Paper ChEl+ ChEI- Placebo Placebo- (A+B) (A+C)

otal = 3377 Avg = 225 STD = 172 59.2 68.304 30.109 199.7 237.9

(B+D) (C+D) Subject

Rt Rc % Events % NOT 0% A/(A+B) C/(C+D) A(A+C) C/(A+C) ADIBC var Tx var C var DIFF

SINGLE GROLJP si2 wi wiRi wi(Ri-r)2 wi2 wi* wi*Ri

R* = 0.80491 var R* = 0.024838

ChEI META DIFF = 0.687238 SE META = 0.042101

LL = 0.60472 UL = 0.769755

SECOND si2

0.003086 0.00 1 153 0.009259 0.000844 0.00 1424 0.001391 0.00 1524 0.0 1 O685 0.005688 5.75 E-05 0.002706 0.0008 19 0.000674 0.000782 0.000704

0.039798 0.002653 0.003 15 1

GROUP wi

479.282 1 867.2437

1 OS 1 184.65 1 703.3348 7 18.9035 656.0976 93.5923 1 1 75 -8242 1739 1.5

369.5084 122 1.47 1 1483 . Z 3 1378.254 1419.518

28 l49.41 1876.627 41 71.275

wiRi wi(Ri-r)2 wi2 wi* wi*Ri

R* = 0.865329 var R* = 0.06043 SE META = 0.064775

LL = 0.48393 1 UL = 0.73785

D IFF TX HOMOG Rt - Rc wi D E F wi*Ri wi(R-R*)2 wi2 IV* i w*iRi

R* = 0.0720 16 var R* = 0.006079

ETA DLFFERENCE = 0,071 141 SE DiFF= 0.024435

LL = 0.023248 UL = 0.1 19034

Dropouts

Paper ChEI+ ChEI- Placebo

KY-1 73 KY-3 38 K Y 4 61 KY-? 37 KY-10 234 KY-11 113 KY-12 174 KY-13 45 KY-13 36 KY-15 58 KY-18 15 KY-19 130 KY-21 116 KY-23 17 KY-24 23 KY-25 325 KY-26 32 KY-27 33 KY-28 132 KY-29 157 KY-30 153

otal = 2002 Avg = 95.3

Placebo- (A+B) (C+D) Subjecr

STD = 78.8 118 16.488 54.29 149.9 92.2848 167.598 66.2 209.9

Rt Rc % Events % NOT 0% Ai(A+B) CI(C+D) A(A+C) C/(A+C) AD/BC 7 var C

0.00096 0.0005

0.00066 0.00043 0.001 66

0.00047 0.00355

0.00044 0.00068 0.00077 0.00273 0.00 125 0.00059 0.001 12 0.00 1 34 0.001 18

0,00033 0.001 55 0.00059 0.00073 0.00047

0.02094

0.00 1 0.00065

var DlFF

0.00 154 0.00 104 0.00098 0.00837

0.0023 0.0008 4

0.0032 1 0.00082 0.00 1 O5 0.00 139 0.00363 0.00208 0.00099 0.00308 0.0039

0.00 164 0.00056 0.00239 0.00092 0.00 128 0.00072

0.03524

0.00 168 0.00 1

SINGLE GROUP si2 wi wiRi wi(Ri-r)2 wi2 IV i* wi*Ri

R* = 0.244893 var R* = 0.025628

ChEI META D F F = 0.26558 SE META = 0.035387

LL = O. 196222

UL = 0.334939

SECOND GROUP si2 wi wiRi wi(K-r)2 wi2 wi* wi*Ri

R* = 0.133778

var R* = 0.004424 Placebo LMETA D E F = 0.153645

SE META = 0.0 15966 I

LL = 0.122352 UL = 0.184939

DFF TX HOMOG k Rt - Rc wi DIFF wi*Ri wi(R-R*)2

R* = 0.0891 12 var R* = 0.006679

ETA DIFFERENCE = 0.10491 3 SE DiFF = 0.019819

LL = 0.066068 UL = 0.143757

Dropouts due to Adverse Events Nt N-Events NX-Events Nc

Paper ChEI+ ChEI- Placebo Placebo- (A+B) (A+C) (B+D) (C+D) Subject

KY- 1 KY-3 KY -3 KY -5 KY -7

KY-10 KY-1 1 KY-13 KY-13 KY-14 KY-15 KY- L 6 KY-18 KY-19 KY-? 1 KY -23 KY-24 KY-25 KY-26 KY-27 KY-28 KY-29 KY-30

Total 1335 5661 225 3156 6996 1560 8817 3381 10377 Avg = 58 246 9.7826 137.22 304.2 67.8261 383.3478 147 451.2 STD = 63.3 148 6.774 66.014 170.4 67.3864 209.775 72 239.1

Rt Rc % Events % Not OR. N(A+B) C/(C+D) A(A+C) C/(A+C) ADiBC var Tx var C var DlFF

SINGLE si2

GROUP wi wiRi wi (Ri- r ) 2 \vi2

R* = 0.132452 var R* = 0.013645

wi* wi*Ri

SE META = 0.024849 LL = O.l2O699 UL = 0.218109

SECOND GROUP si2 wi wiRi wi(Ri-r)2 wi2 wi* wi*Ri

R* = 0.057765 var R* = 0.000195

Placebo META DIFF = 0.059725 SE META = 0.005095

DIFF TX HOMOG k Rt - Rc wi DIFF wi*Ri wi(R-R*)2 wi2 w*i w*iRi

R* = 0.085265 var R* = 0.012168

ETA DEFERENCE = 0.108678 SE DIFF = 0.023955

LL = 0.061725 UL = 0.15563

Quality Analysis - Weighting Eficacy Results by Quality

Cognitive Efficacy - ADAS-cog

Paper

KY- 1 KY-10 KY-1 1 KY-12 KY-35 KY-39 KY-30

Median Quality

3 7 - 2 7 L.

3 3 3

Diff

O. 18567 O. 12449 0.02828 O. 14804 0.0700 1 O. 16907 0.16721

Q2 SQRT

1 WTD AVG = 0.13209 1

Paper

Global Assessrnent Efficacy - CIBIS+ICGIC

Median Q*DIFF Q*wRi Diff QI 42 SQRT UL LL Quality

1 0.09 137 67.3536 0.09137 0.79679 0.888 16 0.03296 0.15597 0.02676 1 0.0700 1 22.2247 0.0700 1 0.63084 0.70085 0.05367 O. 1752 -0.0352 3 - 0.25929 178.353 0.12965 0.67035 0.8 0.03442 0.1971 0.06219 3 0.5 1246 X6.139 0.09 137 0.79679 0.888 16 0.03296 O. 15597 0.02676 3 0.232 12 144.371 0.0700 1 0.63084 0.70085 0.05367 O. 1752 -0.0352 3 0.26784 175.124 0.12965 0.67035 0.8 0.03442 0.1971 0.06219 1 0.22009 31.5491 0.17082 0.64918 0.82 0.0416 0.25336 0.08925 3 O. 2 7947 99.6554 0.07737 0.77888 0.85625 0.03657 O. 14906 0.00569 2 O. 1821 9 176.645 0.08928 0.74926 0.83854 0.03549 0.15883 0.0 1972 3 O. 12455 96.6826 0.22009 0.55932 0.77941 0.08189 0.38059 0.05958

Quality Analysis - Weigh ting Safety Results by Quality

Adverse Events

Paper Median Q*Diff Q*wRi Diff Q1 Q? SQRT UL L L Quality

Dropouts

Paper Median Q*Diff Q*wRi Diff Q1 42 SQRT üL LL Quality

KY- 1 KY -3 KY -4 KY -7 KY-10

KY-1 1 KY-12 KY-13 KY-14 KY-15 KY-18 KY-19 KY -2 1 KY-23 KY -24 KY-75 KY-26 KY-27 KY-28 KY-29 KY-30

Dropouts due to Adverse Events

Paper Median Q*Diff Q*wRi Diff QI Q2 SQRT UL LL Quality

KY-1 KY -3 KY -4 KY -5 KY-7 KY- 10 KY-1 1 KY-12 KY-13 KY-14 KY-15 KY- 1 6 KY-18 KY-19 KY-2 1 KY-23 KY-24 KY-25 KY-26 KY-27 KY-28 KY -29 KY-30

Appendix G - Meta-analysis spreadsheet for single-arrn calculations by ChEI dose

3 z. cc, p- ;O' c. ri SI

High Dose

Paper Success Fail

KY-1 80 70 KY-IO 38 119 KY-11 57 185 KY-21 36 1 09 KY-25 64 1 74 KY-29 30 78

Total =: 314 735 Avg = 52.33 122.5 STD= 16.19 43.76

Total

150 157 242 145 238 117

1049 174.8 47,73

Success rate k A/(A+B) var (SR)

K* - 0.35278

var R* = 0.00138 Mid Dose META AVG = 0.3548

SE META = 0,024 LL = 0.3076 UL = 0.401 9

Higli Dose META AVG = 0.3076 S E M E T A = 0.0416

1,L = 0.2261

U1, = 0.3891 r - Ï

Single Arm Analysis of Global Assessrnent Efficacy (CIBIS+/CCIC) by ChEl dose

Placebo Success rate

Paper Success Fail Total k A/(A+B) var (SR) si2 wi wiRi wi(Ri-r)2 wi2 w i * wi*Ui

Total = 295 1425 1720 10 1.792 12 0.01 005 0.0 1005 12735.5 2082.95 22.2743 18884274 4855.201 8 19 829.97 Avg = 29.5 142.5 172 d f = 0.17921 0.00101 0.00101 1273.55 208.295 1 :I 1 888427 485.5201819 82.99'7

STD = 9.1679 48.63 54.493 9 O.OS 137 0.00063 0.00063 5 16.245 6 1.8723 0.00805 1370306 1 OO.6739Z6 14.92 1

Placebo META AVG = 0.1709 SE META = 0.0144

LL = 0.1428

U L = 0.1991

.- pl: * . CI

3

* . C.

3

CU .- 3

h L . - I$ w . d 3

. I tz . CI

3

. CI

3

PI . - m

C3 > O

via- C\ a C " b - T t w c u o w m k cf T q y , q T :Fii C C , r r C V m D i n a * -- * \ O - M C Z i el b -

R* = 0.24277 var R* = 0.00469

High Dose Success rate

AUTH Success Fail Total k A/(A+B) var (SR) si2 wi wiRi wi(Ri-r)2

KY- 1 KY-1 1 KY-13 KY-14 KY-2 1 KY-25 KY-28 KY-29

Total =

Avg =

STD =

R* - 0.25800 var R* = 0,00406

1.1 E-OS

Mid Dose META AVG = 0.2528 SE META = 0.0308

LL = 0.1923 UL = 0.3132

wi2 wi* wi*Ri

H igh Dose META AVG = 0.2672 SE META = 0.0253

LL = 0.2176 UL = 0.3168

Single Arm Analysis of Adverse Events by ChEI dose

Placebo Success rate

Paper Success

KY -3 KY -4 KY-5 KY-10 KY-I 1 KY-12 KY-13 KY-14 KY-16 KY-18 KY-23 KY-24 KY-28 KY-29 KY-30

Total = 1467 Avg = 97.77 STD = 68.3

Fai 1

7 1 80 16 67 1 O4 47 83 1 O 14 0. 5 54 34 67 45 80

772,s 51.5 30.1 1

Total

1 O6 206 24 239 158 153 164 23 40 93 76 184 274 2 13 286

2239 149.3 83.7

A/(A+B) var (SR) si2 wi wiRi wi(Ri-r)2 wi2 wi* wi*Ri

Plriccbo META AVG = 0.6109 SE META = 0.0648

I L = 0.4839 I J I . = 0.7379

G d - b W W

m i n -

m o n dv, m r u c c E n n

Q g r u - -1 - vi cn O

in In m O 3 Tt- * 00 00 -

Low Dose

Paper Success Fail Total

Total = 147 897 1044 Avg- 24.5 149.5 174 STD = 9.305 61.75 67.22

Success rate k A/(A+B) var

IPlaccbo META AVG = 0.1 529

R* = 0.12818 var R* = 0.00 1 87

SE META = 0.018

LL = 0.1 177 UL = 0.1882

wi2 wi* wiaRi

546452.3 309.925298 1 53.13 407.743 7 1 92.2 1 993 1 26.895 586432.6 166.3066761 2O.28 1 3495 196 190.7656279 27.837 11012581 199.6612303 15.741 630406.6 167.599Oï59 38.308

20348505 1226.477829 182.19 3391418 204.4129716 30.365 37OOl9l 48.8 154741 1 12.35 1

Low Dose META AVG = 0.1 485 SE META = 0.0286

LL = 0.0926 UL = 0.2045

Swccess rate Mid Dose

Paper Success

KY-1 KY -3 KY -4 KY -7 KY-IO KY-13 KY-14 KY-1 5 KY-18 KY-19 KY-26 KY-25 KY-28 KY-29 KY-30

Total =

Avg = 45.33 STD = 34.52

Fai 1

131 99

170 232 66 141 143 215 34 75

1 O6 29

21 1 144 43 1

2227 148.5 96.55

A/(A+B) var wi(Ri-r)2 wi2 wi* wi*Ri

Mid Dose META AVG = 0.2371 SE META = 0.0331

I L = 0.1722 111, = 0.302 1

High Dose S iiccess rate

Paper

KY-1 KY -4 KY-10 KY-1 1 KY-13 KY-14 KY-19 KY-2 1 KY-25 KY-28 KY-29

Total =

Avg =

STD =

Success FaiI Total

157 197 182

243 158 161 148 23 1 418 273 21 1

2379 2 16.3 74.49

A/(A+B) var (SR) si 2 wi wiRi wi(Ri-r)2 wi2 w i * wi*Ri

R* = 0.36263 var R* = 0.06914

Higli Dose META AVG = 0.3705 SE META = 0.0477

LL = 0.277 UL = 0.464

00 O Tl- rn e

C? v) N b

CCI b d

8

Law Dose

Paper Success Fail

Total= 81 1013 Avg = 10.1 3 126.6 STD= 6.712 80.73

Total

105 23

243 27 82

233 24 1 140

1094 136.8 86.9 1

Placebo ETA AVG = 0.0562 SE META = 0.0055

LL = 0.0454 UL = 0.0671

Success rate k A/(A+B) var si2 wi wiRi wi(Ri-r)2

R* = 0,06578 var R* = 0.00037 SE META = 0,0092

LL = 0.0476 1) 1. = 0.0836

Mid Dose

Paper Success

KY-1 9 KY-3 5 KY-4 10 KY-7 30 KY-10 92 KY-13 7 KY-14 13 KY-15 34 KY-18 3 KY-19 57 KY-26 8 KY-28 2 4 KY-29 49 KY-30 46

Total = 387 Avg = 27.64 STD= 24.87

Fai 1

145 1 O4 187 239 84 1 50 153 239 36 92

1 1 1 247 163 506

3456

1 75.4 109.8

Total

154 1 O9 197 269 176 157 166 273 39 149 119 27 1 212 552

2843 203.1 1 15.9

Success rite A/(A+B) var

R* - 0.00406 var K* - 0.00680

Mid Dose ETA AVG = 0.1345 SE META = 0.0232

LL = 0.0891 UL- - 0.1799

High Dose

Paper

K Y - 1 KY -4 KY-IO KY-1 1

KY-1 3 KY-14 KY-19 KY-2 1 KY-28 KY-29

Total =

Avg =

Success Fail

STD = 27.76 39.41

Total

157 196 182 243 158 161 148 23 1 273 21 1

1960 196

40.35

Success rate k A/(A+B) var (SR) si 2 wi wiRi

R* = O. 18202 var R* = 0.02758

Higli Dose ETA AVG = 0.2564 r SE META = 0.0315 LL = 0.1946

UL = 0.3182

Single Arm Analysis of Dropouts due to Adverse Events by ChEl dose

(Studies included in single arm cognitive efficacy analysis only) Placebo

Paper Success

KY- 1 1 1 KY-10 10 KY-11 16 KY-21 39 KY-29 16 KY-30 20

Total = 112 Avg = 18.67 STD = 9.69

Fai I

15 1 1 O7 223 196 197 266

1140 190 50.6

Success rate Total k A/(A+B) var (SR) si2 wi wiRi wi(Ri-r)î

R* = 0.08082 var R* = 0.00066

Placebo META AVG = 0.0854 SE META = 0.01 32

LL = 0.0595 UL = 0.1 113

!2 - = ? - C I -

m r r o d m - -1 m -

o o g C;o's

d

Higb Dose

Paper Success

KY-I 25 KY-IO 104

KY-11 55 KY-21 66 KY-29 67

Total = 317 Avg = 63.4 STD= 25.35

Fai 1

132 78 188 165 1 44

707 141.4

37

Total

157 182 243 23 1 21 1

1024 204.8 3 1 S9

R* = 0,13166

var R* = 0.02262

Success rate k A/(A+B) var(SR) si2 wi wiRi

R* = 0.28804 var R* = 0.03280

Mid Dose META AVG = 0.2399 rl SE META = 0.0684 LL = 0.1 059 UL = 0.374

wi3 wi* wi*Ri

High Dos META AVG = 0.3098 SE META = 0.0759

LL = 0.161 UL = 0.4586

Single Arm Analysis of Dropouts due to Adverse Events by ChEl dose (Studies included in single arm global assessrnent efficacy analysis only)

Placebo

Paper Success Fail Total

Total = 95 1316 1411 Avg= 13.57 188 201.6 STD= 7.519 41.33 48.36

Success rate k A/(A+B) var (SR) si2 wi wiRi

Placebo META AVG = 0.0608 S E M E T A = 0.0113

LL = 0.0387 UL = 0.0829

Low Dose

Paper Success Fail Total

Total = 37 439 476 Avg = 18.5 219.5 238 STD= 0.5 5.5 5

Mid Dose

Paper Success Fail Total

Total = 182 1077 1259 Avg - 30.33 170.5 200.8

STD = 18.51 40.30 45.87

k A/(A-1-B) var si2 wi wiRi wi(Ri-r)2

R* = 0.07757 var R* = 0.00079

k A/(A-tB) var si2 wi wiRi wi(Ri-r)2

1 0.1 1 152 0.00037 0.00037 27 14.8 302.766 0.08833

1 0.04459 0.00037 0.00027 3685.6 1 164.337 l3.8O52 1 0.0783 1 0.00043 0.00043 2399.8 l8O.lO5 1.73605 1 0.32065 0.00 1 18 0.001 18 844.079 370.848 38.9959 1 0.08850 0.0003 0.0003 3357.3 7 307.333 0.9904 1 0.23 1 13 0.00084 0.00084 1 1 O P 275.73 18.7420

wi3 wi* wi*Ri

12552424 1 1 16.476734 82.702 9678338 1069.675403 87,227

22230762 2 186.152 137 169.93 1 1 1 1538 1 1 O!.U.O76068 84.964 1437043 23.40066549 2.2624

SE META = 0.0214 LL = 0.0358 U L = 0.1196

wi2 wi* wi*Ri

R* = 0.10579 var R* = O.OO6 15

High Dose

Paper Success Fail

Total= 266 1011 Avg = 44.33 168.5 STD = 22.28 29

'Total

343 158 161 23 1 273 21 1

1277 2 12.8 4 1.95

Success rate k A/(A+B) var(SR) si2 wi wiRi

Mid Dose META AVG = 0.1412 SE META = 0.0334

LL = 0.0757 UL = 0.2067

Higli Dose META AVG = 0.2023 SE META = 0.0371

L L = 0.1297 UL = 0.275

Appendix H - Meta-analysis spreadsheet for single-arm calcuIations by ChEI class

Single Arm Analysis of Cognitive Efficacy (ADAS-cog) hy ChEI class

First Generation ChEIs Success rate

Paper Success Fail Total k A/(A+B) var si 2 WI wiRi

Total = 21 1 547. 758 3 0.84169 0.00245 0.00245 3786.66 1051.17 Avg = 70.33 182.3 252.7 d f = 0.28056 0.00082 0.00082 1262.22 350.389 STD = 6.944 30.38 36.61 2 0.02003 0.0001 4 0.000 14 224.488 43.6 1 17

R* = 0.2776 var R* = O

wi(Ri-r)2 wi2 wi* wi*Ri

24131454 1559.312052 411.7 1 O:33776 1 O1 6.747562 313.93 1465559 12 10.60273 325.54

4930789 3786,662344 1 O5 1.2 l6&#3596 1262.32078 1 350.39 584323 224.488 1 195 43.612

- 1st Geii META AVE = 0.2776

SE META = 0.0163 LL = 0.2457

- UL = 0.3094

Second Generation ChEIs Success rate

Paper Success Fail Total k A/(A-tB) var si 2 wi wiRi wi(Ri-i-)2 wi2 wi* wi*Ri

KY- 1 137 165 303 1 0.45364 0,00082 0.00082 12 18.47 552.752 30.5057 1484679 78.685 1441 9 35.695

KY-II 93 391 484 1 0.19215 0.00032 0.00032 3118 599.12 33.25 9721936 81.90746814 15.738 KY-21 36 109 145 1 0.24828 0.00129 0.00129 776.918 192.89 1.72637 603601 75.89949424 18.844 KY-29 83 165 248 1 0.33468 0.0009 0.0009 1 1 13.76 372.752 1.7 1693 1 NO469 78.21031433 26,175

KY-30 152 267 419 1 0.36277 0,00055 0.00055 1812.54 657.532 8.22262 3285296 80.38654247 29.162

Total = 501 1097 1598 5 1,59151 0.00388 0.00388 8039.7 2375.04 75.4216 1.6E+O7 395.0889634 125.61 Avg = 100.2 219.4 319.6 df = 0.3 183 0.00078 0.00078 1607.94 475.009 32.;. 196 79.01779167 25.123 STD = 4 1.25 99.82 120.7 4 0.09097 0.00033 0.00033 825.668 170.182 1.6E-15 3347946 2.035 198356 7.1649

2nd Gen META AVE = 0.3 1 79 SE META = 0.0503

LL = 0.2 193 UL = 0.41 65

Single Arm Analysis of Global Assessmen t Efficacy ( C I BIS+/CGIC) by (111 El class

First Ceneration ChEIs

Paper Success Fail

Total = 157 357 Avg= 52.33 119 STD= 18.66 61.6

Success rate Total k A/(A+B) var (SR) si2 wi wiRi

K* = 0.302 var R* = 0.00278

1 st Gen META AVE = 0.3203 SEMETA= 0.0382

LL = 0.2455

UL = 0.3951

Second Generation ChEIs

Paper Success Fail Total

Total = 676 1880 2556 Avg = 96.57 268.6 365.1

STD = 29.38 86.46 101.2

Success rate

k A/(A+B) var si2 wi wiRi

R* = 0.25218 var R * = 0.00516

wi2 wi* wi*Ri

- 2nd Ge11 META AVE = 0.2708

SE META = 0.0286 L L = 0.2147 UL = 0.3269

Single Arm Analysis of Adverse Events by ChEl class

First Generation ChEIs Success rate

Paper Success Fail Total k A/(A+B) var si2 wi wiKi wi(Ri-r)2 wi2 wi* wi*Ri

Total = 764 263 1027 4 2.89444 0.00543 0.00543 10924.2 9426.2

Avg = 191 65.75 256.8 df = 0.72361 0.00136 0.00136 2731.04 2356.55

STD= 160.1 72.07 179.2 3 0.17437 0.00112 0.00112 3316.18 3214.16

R* = 0.86287 var R* = 0.06071

1 st Gcri ETA AVE = 0.7263 S E META = 0.0922

LL = 0.5456 UL = 0.907

Second Generation ChEls

Paper

KY -3 KY-4 KY-5 KY-11

KY-13 KY-14 KY-16 KY-18 KY-28 KY -29 KY-30

Total =

Avg =

STD =

Success

79 273 14

392 230 166 1 1 80 447 390 53 1

2613 237.5 174.5

Fai 1

135 120 9 92 85 161 16 41 97 33 161

950 86.36 52.77

Total

2 14 393 23

484 315 327 27 12 1

544 423 692

3563 323.9 203.6

Success rate k A/(A+B) var (SR) wi

9 18.926 1852.82 96.5635 3 1 43.85 1598.77 1308.3 1 1 1 1.835 540.1 1

37 12.94 5880.88 3876.1 3

23041.1 2094.65 1754.85

R* =

var R* =

wi*

44.7632699 45,89001 18 3 1.6381629 46.36 15498 45.7 1 O1 075 45.4217914 33.1 IC)9926 43 .2844376 46.4665753 46h8 194 19 46.49 10709

475.8289 12 43.257 1738 5.2205 1678

211d Cc11 ETA AVE = 0.6739 SEMETA= 0.0458

LL = 0.584 LIL = 0.7637

Single Arm Analysis of Dropouts by ChEl class

First Generation ChEIs Success rutc

Paper Success Fail Total k A/(AtB) var si2 wî wiRi wi(Ri-r)2 wi2 wi* wi*Ri

Total = 903 783 1686 6 2.71454 0.0071 8 0.00718 745 1.77 3971.17 1 256.535 Q = I 1 .2E+07 714.3040009 269.26 Avg = 150.5 130.5 28 1 df = 0.45242 0.001 2 0.001 2 1241.96 661.867 1 p = 1 1947536 1 19.0506668 44.877 STD = 109.9 56.67 15 1.6 5 0.17224 0.00082 0.00082 636.449 492.972 1 2.2E-53 1 15934 18 70.2447 15 14 28.553

R* = 0.53292 var R* = 0.04275

IstGcn M E T A A V E = 0.377 SE META = 0.0374

LL = 0.3036 UL - 0.4503

Second Generation ChEls Success rate

Paper

KY-1 KY -3 KY-4 KY -7 KY-1 1

KY-13 KY-14 KY-15 KY-18 KY-21 KY-26 KY-27 KY -2 8 KY-29 KY-30

Total =

Avg =

Success

73 28 61 37 I l 3

45 46 58 15 116 32 33 132 157 153

1 O99 73.27

STD = 46.36

Fai 1

238 186 336 232 373 270 28 1 215 1 O6 348 328 146 412 266 539

4276 285.1 106

Total

31 1 214 397 269 486 3 15 327 273 12 1 464 360 179 544 423 692

5375 358.3 143.4

var (SR)

0.00058 0.00053 0,00033 0.00044 0.00037 0.00039 0.00037 0.0006 1 0.0009 0.0004 0.00022 0.00084 0.00034 0.00055 0.00025

0.007 12 0.00047 0.000 1 9

wi

1731.34 1881.79 3052.83 2267.6

2723.46 2572.5 2705.07 163 1.63 1 1 14.19 2474.67 4445.12

1 190.4 2960.23 1812.34 40 18.26

3658 1.4 2438.70 O 14.338

R* L.

var R* -

wi*

194.0476533 195.8022 124 203.9422 132 199.33 lO9Ol 202.3077566 201.4296837 202.2056444 192.7276534 182.7051915 200.808071 7 208,3007699 184.6435764 203.5 169277 l9XQ4635 1 207.268982 1

2974.062062 l98.27O804 1 7.228449497

2ridGcn M E T A A V E - 0.191 r S E M E T A = 0.0183 LL = 0.1551

IJL = 0.227

Single Arm Analysis of Dropouts due to Adverse Events by ChEl class

First Generation ChEIs Success rate

Paper Success Fail Total k A/(A+B) var si2 wi wiRi

KY-IO 196 162 358 1 0.54749 0.00069 0.00069 1445.03 791.136 KY-12 98 289 387 1 0.25323 0.00049 0.00049 2046.49 518.232 KY-19 100 197 297 1 0.3367 0,00075 0.00075 1329.85 447.761 KY-23 13 70 83 1 0.15663 0.001 59 0.001 59 628.337 98.4 143

KY-24 18 60 78 1 0,23077 0,00228 0,00228 439.4 101.4

KY-25 265 214 479 1 0.55324 0.00052 0.00052 1937.97 1072.15

Total = 690 992 1682 6 2.07805 0.00632 0.00632 7827.08 3039.1 Avg = 1 15 165.3 280.3 df = 0.34634 0.00 105 0.00 105 1304.5 1 504.85

STD = 90.73 80.48 151.1 5 0.1535 0,00066 0.00066 602.469 350.046

1 st Gen META AVE = 0.2656 SE META = 0.0297

LL = 0.2074 UL = 0.3238

Second Ceneration ChEIs Success rate

Paper

KY-1 KY -3 KY-4 KY -5 KY -7 KY-1 1 KY-13 KY-14 KY-15 KY-16 KY-18 KY-21 KY-26 KY-27 KY-28 KY-29 KY-30

Total =

Success

34 18 26 0.5 30 73 23 25 34 0.5 10 85 22 19 74 116 55

645

Fai 1

277 196 37 1 23.5 239 413 292 302 129 26.5 1 1 1 379 338 158 470 307 637

4669 274.6

155

Total

31 1 214 397 24

269 486 315

327 163 27 12 1 464 360 177 544 433 693

5314 3 12.6 1 75.9

k A/(A+B) var (SR) si2 wi wiRi

2nd Gcii META AVE = 0.1 O78 SE META -. 0.01 32

1-1.. = 0.082 UL :-. 0.1336