ARS Question #1 ARS Question #2 Which of the following is TRUE about - - PowerPoint PPT Presentation

1

Imaging Neurodegenerative Diseases: When Should We Get Fancy?

Gil Rabinovici, MD

Edward Fein & Pearl Landrith Distinguished Professor UCSF Memory and Aging Center UCSF Recent Advances in Neurology February 14, 2018

Disclosures

• Consulting/research relationships with Avid

Radiopharmaceuticals, Eli Lilly, GE Healthcare, Merck

• Presentation includes the amyloid tracer

[11C]PIB and tau tracer [18F]AV1451 which are not FDA-approved for clinical use

• All case vignettes are based on UCSF

Memory & Aging Center patients, identifying features have been modified

ARS Question #1

Which of the following is TRUE about amyloid PET?

• A. It is investigational only and not yet FDA

approved for clinical use

• B. It is a definitive diagnostic test for

Alzheimer’s disease

• C. Can be clinically useful in patients with

cognitive impairment of uncertain etiology

• D. Amyloid PET ligands bind to soluble

more than fibrillar forms of Aβ

I t i s i n v e s t i g a t i

• n

a l

• n

l y . . . I t i s a d e f i n i t i v e d i a g n

• s

t i . . . C a n b e c l i n i c a l l y u s e f u l i n . . . A m y l

• i

d P E T l i g a n d s b i n d . .

23% 13% 60% 3%

ARS Question #2

Which of the following patients would be highest priority for clinical molecular imaging?

A. Tau PET in an 84 year-old retired attorney with clinically typical Alzheimer’s disease B. Amyloid PET in a 53 year-old commercial airline pilot with early-onset dementia C. FDG-PET in a 45 year-old with subjective memory complaints, normal cognitive testing, and a family history of Alzheimer’s disease D. DaTscan in a 75 year-old with cognitive decline, parkinsonism, and recurrent visual hallucinations

T a u P E T i n a n 8 4 y e a r

• l

d . . . A m y l

• i

d P E T i n a 5 3 y e a r . . . F D G

• P

E T i n a 4 5 y e a r

• l

d . . . D a T s c a n i n a 7 5 y e a r

• l

d . . .

0% 9% 25% 66%

2

Outline

• Limitations of purely clinical diagnosis
• MRI: more than just a “rule out” test
• FDG-PET, DaTscan
• Amyloid PET

– Accuracy vs. neuropathology – Clinical applications – Appropriate Use Criteria – IDEAS: national study on clinical utility

• Preview of Tau imaging

Clinical Diagnosis of AD Circa 1984

McKhann et al., Neurology 1984

“Clinical criteria for AD include insidious onset and progressive impairment in memory and other cognitive functions. The diagnosis cannot be determined by laboratory tests – these tests are important primarily in identifying other causes of dementia”

1984 – 2018: Evolution of AD Biomarkers

FDG PET Structural MRI Amyloid PET CSF Aβ42, Tau, p-Tau

In 2018, diagnosis of AD is still based on clinical symptoms and cognitive testing. Laboratory tests are still used primarily to exclude

• ther causes of dementia.

Accuracy of Clinical Diagnosis of AD

U.S. Alzheimer’s Disease Centers 2005-10

• Probable AD (NINCDS-ADRDA)
• vs. autopsy* (n=526)

– Sensitivity 71%, specificity 71%

• Clinical diagnosis of non-AD dementia
• vs. autopsy* (n=271)

– 39% found to have primary AD post-mortem

* Based on pathologic criteria of intermediate or high-likelihood AD (NIA-Reagan: CERAD-mod or freq; Braak III-VI)

Beach et al., J Neuropathol Exp Neurol 2012

3

Limitations of Clinical Diagnosis

• Same clinical phenotype can be due to

multiple pathologies

• Same pathology can lead to multiple

clinical phenotypes

• Clinical manifestations occur late in

disease course

– Long pre-clinical phase offers a therapeutic window for early intervention

MRI: Not Just a “Rule Out” Test

• Atrophy patterns

– Cortex, medial temporal, basal ganglia, brainstem, cerebellum

• Cerebrovascular

– Infarcts; white matter hyperintensities

• Diffusion abnormalities

– Cortical and subcortical DWI in CJD

• “Pathognomonic” features

– “Hummingbird” – PSP – “Hot cross buns” - MSA

AD FTD

FTD mentioned in radiology report in only 10% of patients

(Suarez et al., Neurology 2009)

Desikan et al. AJNR Am J Neuroradiol. 2013

Clinical Quantitative MRI

FDG-PET in Dementia Syndromes

Bohnen et al. J Nucl Med 2012

4

Diagnostic Accuracy of FDG “AD” vs. “Non-AD”

Sensitivity Specificity

Silverman et al. 2001 (mixed, n=138)

94% 73%

Jagust et al. 2007 (mixed, n=44)

84% 74%

Minoshima et al. 2001 (AD vs. DLB, n=53)

90% 80%

Foster et al. 2007 (AD vs. FTD, n=45)

97% 86%

Lost in Translation?

• Comparison of FDG reads in the community to expert

clinical diagnosis (Shipley et al., Neurol Clin Pract 2013) – FDG performed in 49/1580 (3.1%) – PET reads discordant with final expert dx in 65%

• Sources of error

– Ambiguous metabolic patterns

• Very little use of SSP or other statistical displays

– Limited radiologist experience with brain PET

• 89% read by general radiologists or non-

radiologists – Non-specific or inappropriate clinical question

• 7% ordered for AD vs. FTD

Dopamine Imaging: 123I-FP-CIT SPECT (DaTscan)

• FDA approved for

Parkinson’s disease vs. essential tremor

• Some studies support

utility in DLB vs. AD

• More sensitive than

clinical exam for dopaminergic deficit

• Not useful in identifying

cause of degenerative parkinsonism (e.g. PD

• vs. MSA or PSP)

Normal Abnormal

McKeith et al. Lancet Neurol 2007

Pittsburgh Compound B (PIB) Amyloid plaques

Imaging Amyloid Plaques (11C-PIB PET)

Klunk et al., Ann Neurol 2004

5

NHCH3 N O O O

18F

18F-florbetapir (AmyvidTM)

FDA approved April 2012

18F-flutemetamol (VizamylTM)

FDA approved October 2013

18F-florbetaben (NeuraceqTM)

FDA approved March 2014

Clark et al. JAMA 2011

Pathology Validation: Florbetapir PET

Amyloid PET Visual Reads PET vs. Autopsy Studies

Gold standard: moderate- frequent neuritic plauques (CERAD)

1 – Clark et al., Lancet Neurol 2012 2 – Curtis et al., JAMA Neurol 2015 3 – http://www.accessdata.fda.gov/drugsatfda_docs/label/2014/204677s000lbl.pdf

Tracer N Report Sensitivity Specificity Florbetapir (Amyvid)1 59 Median 92% 95% Flutemetamol (Vizamyl)2 68 Median 88% 88% Florbetaben (Neuraceq)3 82 Median 98% 80%

At a Loss for Words

• 57 year-old RH practicing internist with

2 years of progressive word-finding difficulties

– Struggles to come up with words that should be familiar – Embarrassing socially and professionally – Feels less efficient at accomplishing tasks – No other cognitive or physical symptoms – Has not impacted daily function

6

• General physical and neurological exams wnl
• MMSE 30/30
• Fluent speech with occasional pauses; poor

repetition; comprehension/reading/writing wnl

• Boston Naming Test 12/15 (normal≥14)
• Average performance for age on tests of

executive functions

• Above average on verbal memory, superior

visual memory

At a Loss for Words

• Basic laboratory work-up normal
• MRI: “age-appropriate global volume loss,

mild periventricular white matter changes”

• Patient’s questions:

– Is this normal aging? – If not, what is the diagnosis?

• I’m worried about Alzheimer’s but my memory

is fine – Can I keep working and if so for how long? – Should I take a cholinesterase inhibitor?

At a Loss for Words

At a loss for Words: PET Results

• Diagnosis: MCI – High likelihood due to AD (NIA-AA)

Logopenic-variant primary progressive aphasia

• Treatment

– Cholinesterase inhibitor – Referral to anti-A clinical trial – Early retirement due to medical disability

[11C]PIB-PET [18F]FDG-PET

L R L R

Amyloid Positivity in Normal Older Adults: Concept of Preclinical AD

Mintun 2006; Pike 2007; Mormino 2009 & 2011; Sperling 2011; Chételat 2012; Petersen 2015

15%-30% of cognitively normal

• lder adults are Aβ+
• More common in

ApoE4+ and

• lder age

Aβ+ “controls”

• AD-like structural

and functional brain changes

• Longitudinal

cognitive decline

• Elevated risk of

incident cognitive impairment

7

Aβ PET+ Predicts Cognitive Decline in Aging

Donohue et al., et al., JAMA 2017

Ain’t Misbehavin’

A forgetful veteran

61 yo RHM, 6 years of cognitive and behavioral changes

• First symptoms: misplacing

personal items

• Increasingly repetitive; lost in

familiar environments

• Disinhibited, impulsive,

prone to outbursts

• Overeating, 35 lb weight

gain

• h/o PTSD, EtOH abuse

Riding blind

55 yo LHM, 9 yrs of personality and behavioral changes

• First symptoms: disinhibited

comments to strangers

• New compulsive behaviors

(sorting neighbor’s garbage)

• Skinny dipping at family

beach picnic

• Pretended to be blind so that

dog could ride public transit

Ain’t Misbehavin’

A forgetful veteran

Exam

• Apathetic, slow, perseverative
• Mild limb rigidity

Neuropsych

• MMSE 26/30
• Impaired executive function,

episodic memory, visuospatial

• NPI 52

Differential diagnosis

• Frontal-variant AD vs.

behavioral-variant FTD

Riding blind

Exam

• Jocular, disinhibited
• Mild limb rigidity

Neuropsych

• MMSE 20/30
• Impaired executive function,

episodic memory, visuospatial

• NPI 46

Differential diagnosis

• Frontal-variant AD vs.

behavioral-variant FTD

Differential Diagnosis: AD vs. FTD

• Ideal scenario for amyloid imaging

– Clinical diagnosis is challenging – Aβ a core feature of AD but not part of FTD pathologic spectrum (Tau or TDP-43) – Accurate diagnosis has implications – Early-onset dementia patients less likely to have “age-related” amyloid – Amyloid PET outperforms FDG-PET in this scenario (Rabinovici et al., Neurology 2011)

8

ARS Question #3

Which patient/s do you think will be amyloid PET positive?

• A. Forgetful veteran
• B. “Blind” bus rider
• C. Both
• D. Neither

F

• r

g e t f u l v e t e r a n “ B l i n d ” b u s r i d e r B

• t

h N e i t h e r

52% 12% 24% 12%

A forgetful veteran Riding blind FDG FDG PIB PIB Follow-up

Autopsy: CBD

Follow-up

Autopsy: AD

L R L R L R L R

Amyloid prevalence (%)

100 90 80 70 80 80 60 50 40 30 20 10

AD PCA LPA FTD SD PNFA bvFTD VaD DLB PDD CBS Clinical diagnosis

AD=Alzheimer’s disease; PCA=Posterior cortical atrophy; LPA=Logopenic aphasia; (bv)FTD= (behavioral variant) Frontotemporal dementia; SD=Semantic dementia; PNFA= Progressive non-fluent aphasia; VaD=Vascular dementia; DLB=Dementia with Lewy bodies; PDD=Parkinson’s disease dementia; CBS=Corticobasal syndrome.

N= 1358 54 70 287 62 59 24 138 51 22 56

Ossenkoppele et al., JAMA 2015

Prevalence of Amyloid PET+ in Dementia Syndromes

Positive Predictive Value of Amyloid PET Decreases with Age

Ossenkoppele et al. JAMA 2015

9

Core Criteria:

• Cognitive complaint with
• bjectively confirmed impairment
• Uncertain diagnosis (with AD as a

possibility) after comprehensive evaluation by a dementia expert

• Knowledge of Aβ status expected

to increase diagnostic certainty and alter management

33

Top Clinical Scenarios:

• Persistent/progressive

unexplained MCI

• “Possible” AD

– Atypical or mixed course – Significant co-morbidities (e.g. vascular, psychiatric, substance abuse)

• Atypically early age-of-onset (<65

years)

Johnson et al., Alzheimers Dement/J Nuc Med 2013

Inappropriate Clinical Uses

• Initial evaluation of cognitive complaints

– Scan not a substitute for clinical evaluation

• Screening of cognitively normal individuals

– Pre-clinical AD is a research concept only! – Non-medical use (disability, employment)

• Differentiate AD from other Aβ diseases

– Dementia with Lewy bodies; amyloid angiopathy

• Less added value in straightforward clinical

cases

Johnson et al., Alzheimers Dement/J Nuc Med 2013

• National, open-label study on utility of amyloid PET in

~18,500 Medicare beneficiaries meeting AUC

• Eligible patients referred for PET by dementia experts
• Scans covered by CMS, performed and interpreted locally
• Aim 1: Impact of scan on management plan at 3 months
• Aim 2: Impact on major medical outcomes at 12 months
• The primary hypothesis is that, in diagnostically uncertain

cases, amyloid PET will lead to significant changes in patient management, and this will translate into improved medical outcomes

• Secondary objectives include impact on resource utilization

IDEAS-Study@acr.org IDEAS-Study.org

IDEAS-Study.org

18,295 scans completed Median age 75 (65-105)

60.4% MCI 39.6% dementia

PET Aβ+:

MCI 55.2% Dementia 69.6%

Study closed Jan ’18 Aim 1 results: early 2018 Aim 2 results: mid 2019 592 dementia practices 1,163 dementia experts 343 PET facilities

10

IDEAS-Study.org

Amyloid PET Changes Patient Management (N=3,979)

Change in management composite (%) 25 50 75

Overall MCI Dementia

67.6% 67.8% 65.9%

30% goal Rabinovici et al., AAIC 2017

Composite change in: AD medications, non-AD medications, counseling about safety and future planning

Amyloid PET vs. CSF Aβ42

Florbetapir cortical retention ratio

κ = 0.72

Both positive Both negative

Normal κ = 0.76 EMCI κ = 0.65 LMCI κ = 0.71 AD κ = 0.70

 abnormal

CSF Aβ1-42

normal 

Landau et al, Ann Neurol 2013 Nakamura et al., Nature 2018

Amyloid PET in Drug Development

Sevigny et al., Nature 2016

Aducanumab (humanized monoclonal Anti-Aβ antibody) Phase Ib RCT

11

Landscape of Tau Tracers

Tau PET Across Aging-AD Continuum

Scholl et al., Neuron 2016; Ossenkoppele et al., Brain 2016 Jack et al., Neurology 2016 Schonhaut et al., Ann Neurol 2017

PSP (N=33) vs. NC (46) and PD (26)

12

68 yo retired NFL player with neurobehavioral decline, Aβ-neg

0.0 2.3 AV1451 SUVR

McKee CTE Stage III

Rabinovici et al., HAI 2015

ARS Question #1

Which of the following is TRUE about amyloid PET?

• A. It is investigational only and not yet FDA

approved for clinical use

• B. It is a definitive diagnostic test for

Alzheimer’s disease

• C. Can be clinically useful in patients with

cognitive impairment of uncertain etiology

• D. Amyloid PET ligands bind to soluble

more than fibrillar forms of Aβ

I t i s i n v e s t i g a t i

• n

a l

• n

l y . . . I t i s a d e f i n i t i v e d i a g n

• s

t i . . . C a n b e c l i n i c a l l y u s e f u l i n . . . A m y l

• i

d P E T l i g a n d s b i n d . .

11% 3% 82% 4%

ARS Question #2

Which of the following patients would be highest priority for clinical molecular imaging?

A. Tau PET in an 84 year-old retired attorney with clinically typical Alzheimer’s disease B. Amyloid PET in a 53 year-old commercial airline pilot with early-onset dementia

• C. FDG-PET in a 45 year-old with subjective memory

complaints, normal cognitive testing, and a family history of Alzheimer’s disease

• D. DaTscan in a 75 year-old with cognitive decline,

parkinsonism, and recurrent visual hallucinations

T a u P E T i n a n 8 4 y e a r

• l

d . . . A m y l

• i

d P E T i n a 5 3 y e a r . . . F D G

• P

E T i n a 4 5 y e a r

• l

d . . . D a T s c a n i n a 7 5 y e a r

• l

d . . .

1% 0% 10% 89%

Take Home Points

• MRI can be highly informative in expert hands
• When should we get fancy in 2018?

– When we are diagnostically uncertain – When stakes of diagnosis are high – When test likely to change management – MAY: Mild, Atypical, Young

• Use the right test to answer your clinical question

– Know the limitations of each technique

• Towards era of precision care in neurodegeneration

– From clinical syndromes to molecular phenotyping – From symptomatic therapy to early detection and prevention

13

UCSF-MAC

Bruce Miller Jalayne Arias Nagehan Ayakta Alexandre Bejanin Viktoriya Bourakova Jungho Cha Kiran Chaudhary Leonardo Iaccarino Renaud La Joie Manja Lehmann Orit Lesman-Segev Rik Ossenkoppele Julie Pham Daniel Schonhaut Salvatore Spina Richard Tsai Adrienne Visani Adam Boxer Lea Grinberg Marilu Gorno-Tempini Anna Karydas Joel Kramer Zach Miller Howie Rosen Bill Seeley

Funding

NIA R01-AG045611, P01- AG1972403, P50-AG023501 NINDS U54NS092089 Tau Consortium Michael J. Fox Foundation AFTD Alzheimer’s Association Avid Radiopharmaceuticals American College of Radiology French Foundation

Acknowledgments

UC Berkeley/LBNL

Bill Jagust Suzanne Baker Mustafa Janabi Sam Lockhart Anne Maass Kris Norton Jim O’Neill Michael Scholl IDEAS Study Team Charlie Apgar Maria Carrillo C Gatsonis Ilana Gareen Lucy Hanna Bruce Hillner Cynthia Olson Barry Siegel Rachel Whitmer

Extra Slides

Alzheimer's & Dementia 2015

Aggregated Aβ Lesions

Diffuse Plaques Neuritic Plaques Cerebral Amyloid Angiopathy (CAA)

Slide credit: UCSF Neurodegenerative Disease Brain Bank

14

Thal Phase: spatial extent of Aβ plaques CERAD: density of neuritic plaques Misdiagnosis Impedes Drug Development

Liu et al. Neurology 2015

Amyloid vs. FDG-PET in Differential Diagnosis of AD vs. FTD

Rabinovici et al. Neurology 2011

AD (N=62, age 65, MMSE 22) FTD (N=45, age 65, MMSE 22) Amyloid (PIB) PET visual reads 90% sensitivity, 83% specificity Inter-rater agreement κ=0.96 FDG-PET visual reads 78% sensitivity*, 84% specificity Inter-rater agreement κ=0.72* 95 autopsy-proven dementia cases

PIB: Sensitivity 95%, Specificity 85% FDG: Sensitivity 76%, Specificity 81%

* - p<0.05 vs. PIB