Download - Landau MCI 2014- Final Slides
-
8/13/2019 Landau MCI 2014- Final Slides
1/22
Amyloid, Glucose Metabolism, and Longitudinal Change
Susan Landau
Helen Wills Neuroscience Institute
University of California, Berkeley
Lawrence Berkeley National Lab
-
8/13/2019 Landau MCI 2014- Final Slides
2/22
-
8/13/2019 Landau MCI 2014- Final Slides
3/22
-
8/13/2019 Landau MCI 2014- Final Slides
4/22
Methods
ADNI 1/GO/2 subjects (Normal, Early MCI, Late MCI, AD)
ADNI 1: Up to 9 years of cognitive followup, concurrent FDG & florbetapir
ADNI GO/2: Baseline and 2-year florbetapir & FDG scans
Florbetapir
Cortical retention across Freesurfer-definedtemporal, parietal, cingulate, frontal regions
(composite reference)
FDG
AD-specific MetaROI(pons/vermis reference)
Cognitive
ADAS-cog, AVLT free recall
Longitudinal associations: Mixed effects regression
L Angular
GyrusR Angular
Gyrus
R Inf
Temporal
Gyrus
L Inf
Temporal
Gyrus
Post
Cingulate
Gyrus MetaROI
Landau et al. Neurobiol of Aging 2009
composite
reference region
-
8/13/2019 Landau MCI 2014- Final Slides
5/22
-
8/13/2019 Landau MCI 2014- Final Slides
6/22
Agreement between florbetapir and FDG
FDG metaROI mean
Both
positive
Both
negative
Corticalflorbetap
irmean
Normal
Subj Mem Compl
Early MCI
Late MCI
AD
Total
N265
50
299
217
171
1002
p-value0.05
ns
0.003
-
8/13/2019 Landau MCI 2014- Final Slides
7/22
ADNI timeline
Cognitive testing
FDG
Florbetapir
ADNI 1 Year 0 1 2 3 4 5 6 7
How are florbetapir & FDG status related to longitudinal cognitive decline?
-
8/13/2019 Landau MCI 2014- Final Slides
8/22
Normals N=90
ADAS-cog
ADAS-c
og
-6 -5 -4 -3 -2 -1 0Time (yrs) relative to scan
Florbetapir -
Florbetapir +
FDG -
FDG +
Time (yrs) relative to scan
0.22 pts/yr greater decline 0.50 pts/yr greater decline
-
8/13/2019 Landau MCI 2014- Final Slides
9/22
ADAS-cog
Normals N=90
Time (yrs) relative to scan Time (yrs) relative to scan
FDG - / Florbetapir - FDG - / Florbetapir +
ADAS-c
og FDG + / Florbetapir - FDG + / Florbetapir +
N = 45, 7 ApoE4+
0.0 pts/yr decline
N = 13, 3 ApoE4+
0.3 pts/yr decline
N = 14, 5 ApoE4+
0.6 pts/yr decline
N = 18, 7 ApoE4+
0.2 pts/yr decline
-
8/13/2019 Landau MCI 2014- Final Slides
10/22
MCI N=99
ADAS-cog
ADAS-c
og
Time (yrs) relative to scan Time (yrs) relative to scan
Florbetapir -
Florbetapir +
FDG -
FDG +
1.9 pts/yr greater decline 1.3 pts/yr greater decline
-
8/13/2019 Landau MCI 2014- Final Slides
11/22
MCI N=99
Time (yrs) relative to scan Time (yrs) relative to scan
FDG + / Florbetapir +
ADAS-cog
FDG + / Florbetapir -
ADAS-cog
FDG - / Florbetapir - FDG - / Florbetapir +
N = 22, 3 ApoE4+, 2 converters
0.1 pts/yr decline
N = 16, 2 ApoE4+, 8 converters1.4 pts/yr decline
N = 17, 9 ApoE4+, 3 converters
0.4 pts/yr decline
N = 44, 32 ApoE4+, 31 converters2.5 pts/yr decline
-
8/13/2019 Landau MCI 2014- Final Slides
12/22
So far
Normal aging Amyloid+ status linked to decline
MCI FDG+ status linked to decline
In MCI, combining amyloid and FDG tells us more
about progression than either marker individually
-
8/13/2019 Landau MCI 2014- Final Slides
13/22
ADNI timeline
Cognitive testing
FDG
Florbetapir
ADNI 1
FDG
Florbetapir
ADNI GO/2
Year 0 1 2 3 4 5 6 7
Cognitive testing
Year 0 1 2
What is the relationship between longitudinal florbetapir and FDG
measurements?
-
8/13/2019 Landau MCI 2014- Final Slides
14/22
Florbetapir change over 2 years
CorticalS
UVR
(compositeref)
Age
Normal
N=68
Early MCI
N=84
Late MCI
N=28
AD
N=15
Total N = 195
-
8/13/2019 Landau MCI 2014- Final Slides
15/22
Amyloid change N=195
SUVRannualchange
Baseline SUVR
-
8/13/2019 Landau MCI 2014- Final Slides
16/22
Amyloid change
SUVRannual
change
Baseline SUVR
Villemagne et al. Lancet Neurol (2013)
Villain N et al. Brain 2012
Jack et al. Neurology (2013)
-
8/13/2019 Landau MCI 2014- Final Slides
17/22
L Angular
GyrusR Angular
Gyrus
R Inf
Temporal
Gyrus
L Inf
Temporal
Gyrus
Post
Cingulate
Gyrus
Decreasing AV45 Increasing AV45Annual florbetapir change
Baseline
Florbetapir +
N N=68
EMCI N=84
Baseline
Florbetapir -
FDG
change
Longitudinal florbetapir and glucose metabolism
MetaROI
p=0.03
p=0.02
-
8/13/2019 Landau MCI 2014- Final Slides
18/22
Longitudinal florbetapir and glucose metabolism
N N=68
EMCI N=84
AnnualFDG
SUVR
change Baseline
Florbetapir -
Baseline
Florbetapir +
Voxelwise SPM analysis:
Baseline florbetapir +/-
X florbetapir change
p
-
8/13/2019 Landau MCI 2014- Final Slides
19/22
N
Baseline
Florbetapir +
Increasing florbetapir memory decline
Normals only, regardless of baseline florbetapir status (p = 0.03)
Longitudinal florbetapir and episodic memory change
Decreasing AV45 Increasing AV45Annual florbetapir change
N
EMCI
AVLTcha
nge
Decliningmemory
Im
provingmemory
Baseline
Florbetapir -
-
8/13/2019 Landau MCI 2014- Final Slides
20/22
Summary
Cognitive decline in normal aging linked to amyloid status, decline inMCI linked to FDG
Increasing florbetapir is related to increasing FDG in normal aging andearly MCI; Normals also experience memory decline
Hypermetabolism related to amyloid and AD development has beenreported previously in cross-sectional studies, may be compensatory
(Cohen et al 2009, Oh et al 2012, Ossenkoppele et al 2013, Benzinger et al 2013)
Timing of hypermetabolism relative to other biomarker changes isunclear
Bidirectional FDG changes may complicate detection of metabolicabnormalities early in disease
-
8/13/2019 Landau MCI 2014- Final Slides
21/22
AD
AS-cog
Normals N=90
Time (yrs) relative to scan Time (yrs) relative to scan
FDG - / Florbetapir - FDG - / Florbetapir +
ADAS-c
og FDG + / Florbetapir - FDG + / Florbetapir +
N = 45, 7 ApoE4+
0.0 pts/yr decline
N = 13, 3 ApoE4+
0.3 pts/yr decline
N = 14, 5 ApoE4+
0.6 pts/yr decline
N = 18, 7 ApoE4+
0.2 pts/yr decline
-
8/13/2019 Landau MCI 2014- Final Slides
22/22
Thank you
Michael Weiner
Bob Koeppe
Eric Reiman
Kewei Chen
Norman Foster
Danielle Harvey
Les Shaw
John Trojanowski
Laurel Beckett
Cliff Jack
Chet Mathis
Andrew Saykin
Ron Petersen
Michael Donohue
Anthony Gamst
Art Toga
Karen Crawford
Paul Aisen
ADNI
ADNI participants & staff
UC Berkeley
Bill Jagust
Suzanne Baker
Allison Fero
Cindee Madison