Brain Circuits: Breakdown in Dementia and Hopes for Repair 11/8/12 Gil D. Rabinovici, MD William Seeley, MD Adam Boxer, MD, PhD BIOGRAPHY: Born and raised in Jerusalem, Dr. Rabinovici received his BS degree from Stanford University and MD from Northwestern University Medical School. He completed an internship in internal medicine at Stanford University, neurology residency at UCSF and a behavioral neurology fellowship at the Memory and Aging Center, where he has remained on faculty as an attending neurologist. His research focuses on how structural, functional and molecular brain imaging techniques can be used to improve diagnostic accuracy in dementia and to study the biology of neurodegenerative diseases. Dr. Rabinovici’s work is supported by the National Institute on Aging, the Alzheimer’s Association, the John Douglas French Alzheimer's Foundation and the Hellman Family Foundation. He is the recipient of the 2012 American Academy of Neurology Research Award in Geriatric Neurology, and the 2010 Best Paper in Alzheimer’s Disease Neuroimaging: New Investigator Award from the Alzheimer’s Association.

BIBLIOGRAPHY: Blennow K, de Leon MJ, Zetterberg H. Alzheimer’s disease. Lancet. 2006 Jul 29;368(9533):387-403. Jack CR Jr, Knopman DS, Jagust WJ, Shaw LM, Aisen PS, Weiner MW, Petersen RC, Trojanowski JQ. Hypothetical model of dynamic biomarkers of the Alzheimer's pathological cascade. Lancet Neurol. 2010 Jan;9(1):119-28. Landau SM, Marks SM, Mormino EC, Rabinovici GD, Oh H, O'Neil JP, Wilson RS, Jagust WJ. Association of lifetime cognitive engagement and low β-amyloid deposition. Arch Neurol. 2012 May;69(5):623-29. Rabinovici GD, Jagust WJ. Amyloid imaging in aging and dementia: testing the amyloid hypothesis in vivo. Behav Neurol. 2009;21(1):117-28.

Breakdown of Brain Circuits Part I: Alzheimer’s Disease Gil Rabinovici, MD Assistant Prof of Neurology

The Multidimensional Mind UCSF Mini-Medical School November 8, 2012 UCSF Memory and Aging Center

Sgt. M: The Forgetful Veteran • 70 year-old right-handed veteran with diabetes, high cholesterol – “My memory is terrible”

• Wife notes that in last 3-4 years • • • • •

Forgets conversations, TV programs Repeats questions, stories Memory for remote events spared More quiet in social settings Last month lost on the way home from medical appointment in SF

Image from www.fiu.edu/~weekso/html/amita.htm Helen Wills Neuroscience Institute, UC Berkeley

Sgt. M: Memory Testing

Sgt. M: Memory Testing

After fourth recitation of the list

10 minutes later…

“hat” “cherries” “wrench” “sweater” “lemon” “pliers” “belt” “peaches” “drill”

“hat” “cherries” “wrench” “sweater” “lemon” “pliers” “belt” “peaches” “drill”

Sgt. M: Diagnostic Evaluation Cognitive testing • Moderate-severe memory impairment • Mild impairment in language, visuospatial, executive domains Laboratory testing • Normal electrolytes, liver and kidney function, cell counts, vitamin B12, thyroid function

Brain imaging

Brain Atrophy in Alzheimer’s Disease Temporoparietal cortex • Memory • Language (left) • Math, tool manipulation (left) • Navigation, spatial reasoning (right) Lateral frontal cortex • Executive function Medial frontal cortex spared • Behavior, social function

Rabinovici et al., AJADOD 2008

1

AD Pathology: Plaques & Tangles

The Amyloid Cascade

Amyloid plaques • Extra-cellular • Amyloid-β (Aβ)

Neurofibrillary tangles • Intra-cellular • Tau Courtesy of Roberson & Mucke

Neurofibrillary Tangles

AD Pathophysiology is Complex

Microtubule

Tau protein

Stamelou et al. Brain 2010

Blennow et al., Lancet 2006

Courtesy of Roberson & Mucke

Detecting AD Pathology in Cerebrospinal Fluid

Imaging Amyloid Plaques (PIB-PET) Amyloid plaques

Pittsburgh Compound B (PIB)

• CSF changes in AD – Decrease in Aβ1-42 – Increase in total and phosphorylated tau

• CSF Tau/Aβ1-42 ratio – 85% accurate in discriminating autopsy-confirmed AD from controls Shaw et al. Ann Neurol 2009

Klunk et al. Ann Neurol 2004

2

Clinical Evolution of AD

AD Risk Factors Increase risk • Genes – 1% have disease-causing gene (APP, PS1, PS2) – Apolipoprotein E4 is major risk modifying gene

• Age – Prevalence 1% age 60-64, 35-40% over age 85

• • • •

Female sex Head trauma Vascular risk factors Reduced mental and physical activity

Decrease risk • Education • A little alcohol • Increased mental and physical activity • Heart-healthy diet – Mediterranean

Normal Aging Decline with Age: Processing speed Executive function Naming Memory Improve with Age: Vocabulary General Knowledge Wisdom

Mild Cognitive Impairment

Alzheimer’s Dementia

Decline in memory or other cognitive functions • Beyond expected for age • Does not interfere with day to day function • Multiple causes; may or may not progress to AD

Decline in memory or other cognitive function • Beyond expected for age • Interferes with day to day function

Adopted from: memory.ucsf.edu/Education

Amyloid Deposition in Genetic AD Pre-symptomatic carriers

MCI (median age 44)

Objectives • To investigate in vivo the effects of age-ofonset on distribution and burden of amyloid pathology in AD – Fibrillar A ([11C]PIB-PET)

Dementia (median age 49)

• To compare to effects of age on simultaneously derived functional measures – Resting glucose metabolism ([18F]FDG-PET)

Fleisher et al. Lancet Neurol 2012

PIB in Normal Elderly Correlates with Hippocampal and Cortical Atrophy

Model for AD Cascade

Amyloid-PET CSF Aβ

Mormino et al., Brain 2009

CSF Tau FDG-PET

MRI atrophy Cognitive symptoms

Functional decline

Oh et al., Neuroimage 2010 Jack et al., Lancet Neurol 2010

3

Exercise is Associated with Low Amyloid Burden

Cortical PIB

High Cognitive Activity is Associated with Low Amyloid Burden

Lowest cog activity Alzheimer’s patients

Moderate cog activity

Highest cog activity

Older controls

Young controls

Liang et al., Ann Neurol 2010

Landau et al., Arch Neurol 2012

Take Home Messages • AD attacks brain networks involved in memory, language, visuospatial and executive function • Misfolded proteins form toxic plaques and tangles 10-15 years before symptoms • Increasing focus on disease prevention: – Mens sana in corpore sano • Future of therapy wll include early detection and intervention with biologically specific agents

Bruce Miller Bill Jagust

Acknowledgments

Adi Alkalay Nick Block Adam Boxer Brendan Cohn-Sheehy Mary DeMay Michael Geschwind Pia Ghosh Marilu Gorno-Tempini Lea Grinberg Robin Ketelle Joel Kramer Baber Khan Suzee Lee Manja Lehmann Cindee Madison Katya Rascovsky Kate Rankin Howie Rosen Bill Seeley Trishna Subas Marissa Urbano Mike Weiner Teresa Wu

Funding sources: NIA K23-AG031861 NIA P01-AG1972403 NIA R01 AG027859 ADRC P50 AG023501 CA DHS 04-33516 Alz Association French Foundation Hellman Foundation

4

Predicting regional neurodegeneration from the healthy brain functional connectome

Take home messages •

The human brain is composed not of isolated and specialized brain regions but of large-scale distributed networks

William W. Seeley, MD Associate Professor of Neurology UCSF

•

Network science and brain imaging has provided new tools to examine neural networks in humans

Alzforum Webinar

•

Neurodegenerative diseases represent organized, networkbased degenerations

•

A network-based approach may allow us to predict and follow a patient’s trajectory

April 10, 2012

Network hypothesis of Alzheimer’s disease

CB Saper, BH Wainer, & DC German. Axonal and transneuronal transport in the transmission of neurological disease: potential role in system degenerations, including Alzheimer’s disease. 1987. Neuroscience 23(2): 389-98.

Braak et al, Acta Neuropath 2006

Patient F.T.

Network hypothesis of Alzheimer’s disease Large-scale “default mode network”

58 y.o. business executive with 2 high school children rmPFC

Brought in by wife for increasingly uncharacteristic behaviors:

MTL MTG ANG PCC

PreCu

• • • •

Disinterest in kids’ school and sports activities Speaking out of turn, commenting on strangers’ weight or hairstyle Circles the kitchen island 3 times (counterclockwise) upon entering room New penchant for sweets; overeating in general

Language, memory, navigation, skilled movements all normal. Denies low mood, sleep disturbance, life stressors. Greicius, PNAS 2003, PNAS 2004

1

FRONT

Patient F.T. Anterior Cingulate Cortex (ACC) and Rostromedial PFC

LEFT

RIGHT

BACK

Right Frontoinsula (FI)

R

L

FTD Prevalence

Macedo et al., Behavioral Neurology of Dementia, 2009

Intrinsic connectivity measured with fcMRI DLPFC

Common cause early age-of-onset dementia • Prevalence ~1/5000 in persons 45-64 years old, 1:1 with AD (Ratnavalli et al., Neurology 2002) • Higher incidence than AD when symptoms begin before age 60 years (Knopman et al., Neurology 2004) • Broader FTD spectrum even more common

In healthy subjects, baseline low frequency fMRI BOLD signal fluctuations in Right FI are correlated with…

Frontal pole

Less common after 70?

Lat OFC

Single subject

Left

FI

Right FI seed ROI

ACC

Time (sec)

FI

SLEA VSP

Right

Hypothalamus

“Salience Network” (Intrinsic connectivity network) Seeley et al J Neurosci 2007

3T fcMRI 19 healthy controls

Alzheimer’s disease

bvFTD

Corticobasal Syndrome

Behavioral Variant Frontotemporal Dementia

bvFTD atrophy pattern VBM, patients vs. controls N = 24 Functional connectivity Right FI seed fMRI, healthy controls N = 19

Progressive Nonfluent Aphasia

Semantic Dementia

2

Alzheimer’s Disease

Behavioral variant FTD

Semantic Dementia

bvFTD

AD

SD

Vulnerable Cortical Epicenters? AD: ANG

CBS: R preMC

bvFTD: R anterior insula

Corticobasal Syndrome

Progressive nonfluent aphasia

Network-based neurodegeneration

CBS

PNFA: L IFG oper

SD: L temp pole

PNFA

Lingering questions

• Does each disease involve a focal “epicenter” from which disease spreads? • What are the mechanisms of network-based vulnerability? Time (sec) Single subject

• Can we use network-based imaging in the clinic?

Chicago O’Hare Airport

3

bvFTD bvFTD atrophy pattern VBM, patients vs. controls N = 24

Does each disease involve a focal “epicenter” from which disease spreads?

bvFTD pattern

Correlation between functional connectivity in health and disease vulnerability AD pattern

bvFTD pattern

SD pattern

AD pattern

PNFA pattern CBS pattern

bvFTD pattern

SD pattern

PNFA pattern

CBS pattern

4

Graph metrics in health predict atrophy severity across target and off-target networks PNFA pattern

CBS pattern

AD pattern bvFTD pattern

SD pattern

PNFA pattern

CBS pattern

Atrophy severity in disease

SD pattern

Atrophy severity in disease

AD pattern bvFTD pattern

Graph metrics in health predict atrophy severity across target and off-target networks

Healthy connectivity graph

Healthy connectivity graph

E

Graph metrics in health predict atrophy severity across target and off-target networks PNFA pattern

CBS pattern

AD pattern bvFTD pattern

SD pattern

PNFA pattern

CBS pattern

Atrophy severity in disease

SD pattern

Atrophy severity in disease

AD pattern bvFTD pattern

Graph metrics in health predict atrophy severity across target and off-target networks

Healthy connectivity graph

Healthy connectivity graph

E

E

Graph metrics in health predict atrophy severity across target and off-target networks SD pattern

PNFA pattern

CBS pattern

AD pattern bvFTD pattern

SD pattern

PNFA pattern

CBS pattern

Atrophy severity in disease

Atrophy severity in disease

AD pattern bvFTD pattern

Graph metrics in health predict atrophy severity across target and off-target networks

Healthy connectivity graph

E

Healthy connectivity graph

E

E

E

E

time

5

Salience Network breakdown and DMN enhancement track bvFTD severity

Salience Network breakdown and DMN enhancement track bvFTD severity • Right frontoinsular (FI) dysconnectivity to Salience Network tracks functional impairment • Suitable biomarker for drug development? Needs to predict later clinical benefit

Right

Right

Clinical outcome

Drug Placebo

fMRI 1mo

3mo

6mo

12mo

Acknowledgments Seeley Lab Stephanie Gaus Christine Guo Alex Larkin Norbert Lee Alissa Nana Li Manu Sidhu Andrew Trujillo Formerly: Danielle Carlin Maria Cobos (MGH) Rich Crawford Raquel Gardner Stathis Gennatas (PENN) Eun-Joo Kim (Pusan) Helen Zhou (Duke-NUS, Singapore)

Liang et al J Nsci 2011

UCSF Memory & Aging Center Adam Boxer Marilu Gorno-Tempini Suzee Lee Bruce Miller Howard Rosen Virginia Sturm Michael Weiner

UCSF ADRC Pathology Core Ben Arevalo Kelly Creighton Steve DeArmond Lea Grinberg Eric Huang Jian Yang Jakc Whittembury

Funding Sources: National Institute on Aging Alzheimer’s Drug Discovery Foundation Association for Frontotemporal Dementia Larry L. Hillblom Foundation James S. McDonnell Foundation John D. French Alzheimer’s Foundation UCSF Consortium for FTD Research Tau Consortium Hellman Family Foundation

Cal Tech John Allman MSSM Patrick Hof Barrow Neurological A.D. (Bud) Craig UCI Brain Bank Elizabeth Head Stanford University Michael Greicius Vinod Menon

6

Frontotemporal degeneration (FTD) Treatment Advantages

Current Management of AD Cognitive complaint (patient or family) Investigations (cog. testing, brain image, labs [vitamin B12, thyroid function]) Diagnosis (exclusion of other causes): Mild Cognitive Impairment (no functional decline) Alzheimer䇻s Disease (functional decline)

Adam L. Boxer, MD, PhD

Interventions (manage symptoms): Acetylcholinesterase inhibitor (Aricept®, Exelon® or Razadyne®) and/or Memantine (Namenda®) Exercise (physical/mental) Minimize CV risk factors Minimize other CNS drugs Treat behavioral/psychiatric symptoms Caregiver support

Director, Neuroscience Clinical Research Unit Director, AD and FTD Clinical Trials Program Associate Professor of Neurology University of California, San Francisco UCSF Mini Med School; November 8, 2012

End of Life Care

Current AD Therapies

14

300

12 10

200

Delay onset by 2 years

8 6

Delay onset by 5 years

2003

2010

2020

2030

Year Memory and Aging Center

2040

100 2050

Cost to society: $ billions

Millions of people with AD in the U.S.

•Donepezil (Aricept) •Rivastigmine (Exelon) •Galantamine (Razadyne) •Memantine (Namenda)

Impact of Alzheimer䇻s Treatment

Alzheimer䇻s disease under the microscope: plaques and tangles

Alzheimer’s vs. Mouseheimer䇻s Dude, where䇻s my cheese?

Univ Utah

Amyloid plaques • Introduce human dementia causing genes (tau, APP, PS1) John Trojanowski

Neurofibrillary tangles

Targets for disease modifying agents

Anti-amyloid vaccines antibodies aggregation inhibitors

R-fluribuprofen gamma secretase inhibitor

• Memory deficits, amyloid plaques but no neurofibrillary tangles

Bapineuzumab removes amyloid

beta secretase inhibitor

Bap

Pbo

Prevention of AD: Early Detection + Disease Modification

Too late or wrong target? current anti-amyloid clinical trials

Brain disease

tau amyloid pre-clinical detection disease-modifying treatment

anti-tau drugs

Time Cognitively Normal

Mild Cognitive Impairment (MCI)

Dementia (Alzheimer䇻s Disease) National Institute on Aging

Frontotemporal dementia • Common early onset ( in AD > in FTD

Tartaglia et al., submitted

Memantine and dementia • FDA/EMA approved for mod-severe Alzheimer’s o Not efficacious mild AD (Schneider, 2011) o Better agitation/aggression, eating/appetite, irritability/lability (+donep; Cummings, 2006)

• Case reports in FTD (+/-) • Open label study

• Global, behavioral benefits in PDD/DLB o Aarsland, 2009; Emre, 2010

• Small cog benefit VaD? Boxer, ADAD, 2009

o Kavirajan, 2007

Memantine in bvFTD, svPPA

Multicenter, 26 week, randomized, double blind, placebo controlled trial memantine 20 mg daily in bvFTD and svPPA

• Nine academic centers • Primary outcomes: NPI, CGIC • Secondary: CDR-sb, FAQ, FTD neuropsych battery (Knopman, 2008), TEFA, ZBI, EXIT25 • Included bvFTD and svPPA with characteristic neuroimaging; MMSE > 15 • AChI, memantine, antipsychotics within 1 month excluded

Transient improvement in NPI (ITT)

Assessed for eligibility (n=100)

-8 P=.31

•Diagnosis does not meet criteria (n=2) •Lab abnormality (n=1) •MMSE not in range (n=6) •Exclusionary medication (n=3) •Imaging not consistent with FTD (n=3) •Exclusionary comorbid neurological condition (n=1) ¨ Declined to participate (n=3)

Randomized (n=81)

Allocated to Memantine (n=39)

Allocated to Placebo (n=42)

Least Squares Mean Change (SE)

P=.01 Excluded (n=19) ¨ Not meeting inclusion criteria (n=16)

-6 P=.36

-4 -2 0 2 4 6 Placebo

Memantine

8 Discontinued intervention (n=2) due to adverse event

Completed 26 weeks treatment (n=37)

Discontinued intervention (n=3) ¨ Started excluded medications (Aricept, Namenda) (n=2) ¨ Adverse event (n=1)

Completed 26 weeks treatment (n=39)

Baseline

6

12 Treatment Week

26

Trend towards worse ADLs (FAQ)

Worse digit symbol substitution (no learning effect?) 8

-1 Least Squares Mean Change (SE)

0 1 P=.15

2 3 4

P=.001

6 4

P=.14 P=.69

2 0 -2 -4 -6

5

Placebo

Placebo

Baseline

Baseline

Memantine

-8

Memantine

6 12

6

12

26

Treatment Week

26

Treatment Week

Least Squares Mean Change (SE)

Least Squares Mean Change (SE)

P=.04

Worse Boston Naming

Memantine bvFTD svPPA summary

3

• Small improvement behavior 6 weeks. Underpowered to detect NPI effects? • No difference in CGIC • Trend towards worse ADLs (FAQ) • Worse processing speed (DSST), naming (BNT) • No effect on CDR-SB decline (twice AD rate) • Memantine not useful in FTD

P Rodney Pearlman, Ph.D.

• 4RTNI •

: Howie Rosen, Carmela Tartaglia, Brad Dickerson, Norbert Schuff, Mike Weiner, Les Shaw, John Trojanowski, Anna Karydas, Art Toga, Paul Aisen, Jere Meredith Support: NIA, Tau Consortium, Association for FTD, Alzheimer’s Drug Discovery Foundation, BMS, Envivo, Hellman Foundation