doi:10.

1093/brain/awx320 BRAIN 2018: 141; 271–287 | 271

Widespread brain tau and its association with

ageing, Braak stage and Alzheimer’s dementia
Val J. Lowe,1 Heather J. Wiste,2 Matthew L. Senjem,1,3 Stephen D. Weigand,2
Terry M. Therneau,2 Bradley F. Boeve,4 Keith A. Josephs,4 Ping Fang,1 Mukesh K. Pandey,1
Melissa E. Murray,5 Kejal Kantarci,1 David T. Jones,1,4 Prashanthi Vemuri,1
Jonathan Graff-Radford,4 Christopher G. Schwarz,1 Mary M. Machulda,6
Michelle M. Mielke,2,4 Rosebud O. Roberts,2 David S. Knopman,4 Ronald C. Petersen4 and
Clifford R. Jack, Jr1

See Herholz (doi:10.1093/brain/awx340) for a scientific commentary on this article.

Autopsy data have proposed that a topographical pattern of tauopathy occurs in the brain with the development of dementia due
to Alzheimer’s disease. We evaluated the findings of tau-PET to better understand neurofibrillary tangle development as it is seen in
cognitively unimpaired and impaired individuals. The evolution of Alzheimer’s disease tauopathy in cognitively unimpaired indi-
viduals needs to be examined to better understand disease pathogenesis. Tau-PET was performed in 86 cognitively impaired
individuals who all had abnormal amyloid levels and 601 cognitively unimpaired individuals. Tau-PET findings were assessed
for relationships with clinical diagnosis, age, and regional uptake patterns relative to Braak stage. Regional and voxel-wise analyses
were performed. Topographical findings from tau-PET were characterized using hierarchical clustering and clinical characteristic-
based subcategorization. In older cognitively unimpaired individuals (550 years), widespread, age-related elevated tau signal was
seen among those with normal or abnormal amyloid status as compared to younger cognitively unimpaired individuals (30–49
years). More frequent regional tau signal elevation throughout the brain was seen in cognitively unimpaired individuals with
abnormal versus normal amyloid. Elevated tau signal was seen in regions that are considered high Braak Stage in cognitively
unimpaired and cognitively impaired individuals. Hierarchical clustering and clinical characteristic-based categorizations both
showed different patterns of tau signal between groups such as greater tau signal in frontal regions in younger onset
Alzheimer’s disease dementia participants (most of whom had a dysexecutive clinical presentation). Tau-PET signal increases
modestly with age throughout the brain in cognitively unimpaired individuals and elevated tau is seen more often when amyloid
brain accumulation is present. Tau signal patterns in cognitively unimpaired correspond to early Braak stage but also suggest
tangle involvement in extra-medial temporal and extra-temporal regions that are considered more advanced in the Braak scheme
even when amyloid negative. Our findings also suggest the possibility of widespread development of early tangle pathology rather
than a pattern defined exclusively by adjacent, region-to-region spread, prior to onset of clinical symptoms. Distinct patterns of
neurofibrillary tangle deposition in younger-onset Alzheimer’s disease dementia versus older-onset Alzheimer’s disease dementia
provide evidence for variability in regional tangle deposition patterns and demonstrate that different disease phenotypes have
different patterns of tauopathy. Pathological correlation with imaging is needed to assess the implications of these observations.

1 Department of Radiology, Mayo Clinic, Rochester, MN, USA

2 Department of Health Sciences Research, Mayo Clinic, Rochester, MN, USA
3 Department of Information Technology, Mayo Clinic, Rochester, MN, USA
4 Department of Neurology, Mayo Clinic, Rochester, MN, USA
5 Department of Neuroscience, Mayo Clinic, Jacksonville, FL, USA
6 Department of Psychiatry and Psychology, Mayo Clinic, Rochester, MN, USA

Received July 26, 2017. Revised October 2, 2017. Accepted October 17, 2017. Advance Access publication December 8, 2017
ß The Author (2017). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved.
For Permissions, please email: journals.permissions@oup.com
272 | BRAIN 2018: 141; 271–287 V. J. Lowe et al.

Correspondence to: Val J. Lowe, MD

Department of Radiology, CH 1-285
Mayo Clinic
200 First Street SW, Rochester, MN 55905
E-mail: vlowe@mayo.edu

Keywords: dementia; tau-PET; Alzheimer’s disease; mild cognitive impairment; amyloid-PET

Abbreviations: ADD = Alzheimer’s disease dementia; AUROC = area under the receiver operating curve; MCI = mild cognitive
impairment; PVC = partial volume correction; SUVr = standardized uptake value ratio

development. Therefore, in this study we evaluated the dis-

Introduction tribution of tau-PET findings in cognitively unimpaired
Neurofibrillary tangle topographic distribution in the brain and cognitively impaired individuals. There were four
is the basis for Braak neurofibrillary tangle pathological main aims: (i) assess tau-PET regional uptake as a function
staging of Alzheimer’s disease (Braak and Braak, 1991). of age and other demographics and its ability to distin-
Braak neurofibrillary tangle stage is strongly associated guish between cognitively unimpaired and cognitively im-
with cognitive impairment (Braak and Braak, 1991; paired individuals; (ii) determine the influence of amyloid
Duyckaerts et al., 1997; Bennett et al., 2004; Sabbagh status on regional uptake patterns among cognitively un-
et al., 2010; Nelson et al., 2012) but shows considerable impaired; (iii) assess the similarities of tau-PET findings
diagnostic overlap with cognitively unimpaired individuals with predicted Braak neurofibrillary tangle stage; and (iv)
in autopsy data (Gertz et al., 1996). Some autopsy data of use clinical groups and clusters of individuals to assess
cognitively unimpaired elderly subjects show that neurofib- different tau-PET uptake patterns.
rillary tangle pathology is largely confined to the entorhinal
and adjacent temporal isocortices and not often seen in
temporal neocortex or extra-temporal regions (Braak and
Braak, 1991, 1997; Arriagada et al., 1992; Bouras et al.,
Materials and methods
1994). Quantification of tau burden during life using PET
and 18F-AV-1451 is now possible (Fawaz et al., 2014; Subjects
Hashimoto et al., 2014; Zimmer et al., 2014) and initial Participants were part of the Mayo Clinic Study of Aging or
studies show that AV-1451-PET can identify Alzheimer’s Mayo Clinic Alzheimer’s Disease Research Center (Roberts
disease neurofibrillary tangles (Xia et al., 2013a). Recent et al., 2008). The Mayo Clinic Study of Aging is a rando-
studies have shown that tau-PET closely mimics Braak mized, population-based, ageing study focused on non-
neurofibrillary tangle staging in Alzheimer’s disease demented individuals and encompasses a wide age range.
(Johnson et al., 2016; Ossenkoppele et al., 2016; Schwarz The Mayo Clinic Alzheimer’s Disease Research Center is a
et al., 2016). In normal ageing, tau-PET signal has been clinic-based study. There were 601 cognitively unimpaired
participants and 86 cognitively impaired participants who
described to occur in the medial temporal lobe, but only
completed imaging studies consecutively between 7 April
in other regions when amyloid is present (Scholl et al.,
2015 and 15 February 2017 that were selected. Tau-PET,
2016). amyloid-PET and MRI scans were performed in all partici-
Amyloid-PET can identify amyloid-b plaques in the brain pants. The participants were determined to be cognitively un-
(Klunk et al., 2004; Johnson et al., 2007; Clark et al., impaired by a consensus diagnosis (this includes quantitative
2012; Driscoll et al., 2012; Kantarci et al., 2012) and is data as well as clinical and cognitive assessment by neurolo-
an important tool for understanding disease pathogenesis gists, geriatricians, neuropsychologists, and study coordin-
and for selecting participants with abnormal amyloid-b for ators). The cognitively impaired individuals were selected if
therapeutic trials (Sperling et al., 2014). Soluble oligomeric they had amnestic mild cognitive impairment (MCI) or ADD.
amyloid-b is hypothesized to be a possible cause of tau The diagnosis of MCI was based on published criteria
hyperphosphorylation and the development of neurofibril- (Petersen, 2004). A diagnosis of dementia was based on the
lary tangles (Zheng et al., 2002) but the spatial distribu- Diagnostic and Statistical Manual of Mental Disorders
(American Psychiatric Association, 2000). To increase the
tion of amyloid and neurofibrillary tangles differ
likelihood that the cognitively impaired participants were
(Okamura et al., 2014). A better understanding of the re-
on an ADD pathway, they were required to have abnormal
lationship between tau-PET and amyloid-PET across the amyloid. In addition, clinical notes for all cognitively im-
spectrum from cognitively unimpaired to cognitively im- paired participants were reviewed and participants in which
paired is needed. This study is designed to provide insight an ADD pathway was still unclear (i.e. those without an
on how neurofibrillary tangle pathology might develop in amnestic component to their presentations) were excluded.
dementia due to Alzheimer’s disease (ADD) and how amyl- After exclusions, 86 participants remained in the cognitively
oid and neurofibrillary tangles interact in early disease impaired group. No adverse events were seen from imaging.
Tau imaging in ageing, Braak stage and dementia BRAIN 2018: 141; 271–287 | 273

All participants or designees provided written consent with aged 30–49 (all normal amyloid); the 95th percentile was esti-
approval of Institutional Review Boards. mated from quantile regression where regions were ordered by
median SUVr. We calculated the percentage of individuals with
elevated tau-PET for each individual region of interest within
Neuroimaging cognitively unimpaired abnormal amyloid and normal amyloid
For tau-PET, participants were injected with 370 MBq (range individuals aged 50 years or older and cognitively impaired
333–407 MBq) of 18F -AV-1451 prior to imaging and imaging amyloid abnormal individuals. Logistic regression was used
was performed as four, 5-min frames for a 20-min PET acqui- to determine if the proportion of cognitively unimpaired indi-
sition, 80–100 min post-injection. Amyloid-PET imaging was viduals with elevated tau-PET was different among the abnor-
performed using Pittsburgh compound B and consisted of mal amyloid compared to the normal amyloid individuals,
four 5-min dynamic frames acquired 40–60 min after injection with and without adjusting for age. Permutation tests were
of 628 MBq (range 385–723 MBq) of 11C-Pittsburgh com- used to determine significant associations after accounting for
pound B (PIB) as previously described (Lowe et al., 2014). multiple comparisons. We also defined a group of individuals
MRI scans at 3 T with a 3D volumetric T1 magnetization-pre- with intermediate amyloid levels, using the upper tertile of the
pared rapid gradient-echo sequence were performed as previ- normal amyloid group (amyloid SUVr of 1.33–1.42, n = 142),
ously described (Murray et al., 2015; Jack et al., 2017). to assess the percentage of individuals with elevated tau-PET
signal. Regions with elevated tau-PET signal were compared to
Image analysis regions in Braak staging.
We used agglomerative hierarchical clustering with Ward’s
Cortical regions of interest were defined by an in-house version minimum variance method to separately cluster cognitively un-
of the automated anatomic labelling atlas (Tzourio-Mazoyer impaired (530) and cognitively impaired individuals with
et al., 2002) as previously described (Vemuri et al., 2008a). similar regional tau-PET findings. Based on the clustering den-
Non-linear registration using SPM5 (Ashburner and Friston, drograms, a graphical summary of the dissimilarity between
2005) was used to apply the atlas to each subject’s MRI. clusters, we chose to group individuals into four cognitively
The static tau-PET and amyloid-PET volumes were co-regis- unimpaired and three cognitively impaired clusters by arbitrary
tered to the subject’s own MRI scan. Statistics on image voxel visual selection of groupings. Our preliminary findings from
values were extracted from each labelled cortical region of the hierarchical clustering suggested different topographical
interest. Individual tau-PET region of interest median values
patterns of tau-PET signal. Overall differences in participant
were normalized to cerebellar crus (bilateral crus, 1–2) to cal-
characteristics across the four cognitively unimpaired clusters
culate regional standardized uptake value ratio (SUVr). The
and the three cognitively impaired clusters were assessed with
crus region was selected to provide cerebellar grey matter in
Kruskal Wallis tests for continuous variables and chi-square or
relative isolation from CSF spaces and to avoid adjacency to
Fisher’s exact tests for categorical variables. Formal tests of
parahippocampal, fusiform and lingual gyri to avoid bleed-in
differences in tau-PET among the clusters were not performed
signal from tau-pathology. Data with and without partial
since the clusters were defined by tau-PET. Since the number
volume correction (PVC) using the two-compartment method
of clusters selected was somewhat arbitrary, we also analysed
(Meltzer et al., 1999) were evaluated. Global cortical amyloid-
regional tau-PET findings in a priori subcategorizations of the
PET SUVr was computed from a meta-region of interest nor-
malized to the cerebellar crus where normal or abnormal participants into groups based on clinical diagnosis, amyloid-
status was based on a cut-point of 1.42 (Jack et al., 2017). PET status and age as: cognitively unimpaired, normal amyl-
oid 550 years; cognitively unimpaired, normal amyloid 550
years; cognitively unimpaired, abnormal amyloid 550 years;
Statistical methods amnestic MCI, abnormal amyloid 550 years; ADD, abnormal
The area under the receiver operating characteristic curve was amyloid 560 years, ADD, abnormal amyloid 60–70 years;
calculated to determine group-wise discrimination performance and ADD, abnormal amyloid 570 years.
of tau-PET between cognitively unimpaired and cognitively To qualitatively assess topographic tau-PET findings, images
impaired individuals and between cognitively unimpaired indi- from participants within each cluster were summed and nor-
viduals with normal amyloid versus abnormal amyloid. malized to produce cluster composite images. Region of inter-
Associations between regional tau-PET SUVr and age were est analyses were performed on clusters and clinical
assessed using Spearman rank correlations within cognitively subcategories. Voxel-wise analyses using SPM5 were also per-
unimpaired individuals with normal and abnormal amyloid, formed for clusters and clinical subcategories. Pairwise differ-
and within cognitively impaired individuals for tau-PET ences between cluster groups and clinical subcategories were
SUVr with and without PVC. Partial correlations were used assessed using multiple regression analyses in SPM5 and T-
to assess the associations between tau-PET and age among all statistic differences between pairs of groups. The results were
cognitively unimpaired individuals, after adjusting for amyloid- displayed at a false discovery rate corrected P-value threshold
PET (using continuous SUVr and without PVC). Correction of P 5 0.05 and no cortical region masking or cluster-based
for multiple comparisons across many regions was performed thresholding was performed. The T-statistic group difference
using permutation-based resampling. For a two-sided family- maps were transferred to MNI space using spatial normaliza-
wise type 1 error of 0.05 across 47 regions, we used a critical tion (Avants et al., 2008) and from the structural abnormality
value of 2.8 (rather than the usual 1.96) for testing and con- index (STAND) custom template space to MNI space (Vemuri
fidence intervals. et al., 2008b) and the resulting T-statistic maps were visualized
We defined elevated tau-PET in each region based on the on a 3D rendering with BrainNet-Viewer software (http://
95th percentile among 98 cognitively unimpaired individuals www.nitrc.org/projects/bnv/) (Xia et al., 2013b).
274 | BRAIN 2018: 141; 271–287 V. J. Lowe et al.

regions showed a positive age relationship with tau-PET

Results SUVr. There were 41/47 regions significantly correlated
Cognitively unimpaired and cognitively impaired groups with age in the normal amyloid cognitively unimpaired
differed by age, APOE genotype, Mini-Mental State group with PVC and 20/47 regions significant without
Examination (MMSE), and tau-PET SUVr (Table 1). PVC. Age correlations were generally lower without PVC
Individuals in the cognitively impaired group were required but still significant in several regions. In the cognitively
to have abnormal amyloid-PET scans. impaired individuals, higher tau-PET SUVr was associated
with younger age in most regions. This is likely due to
higher tau-PET signal in younger-onset ADD. The putamen
Group discrimination and pallidum region values increased with age in all groups
Boxplots of tau-PET by clinical diagnosis showed excellent and are considered ‘off-target’ binding (Lowe et al., 2016).
separation between cognitively unimpaired and cognitively
impaired individuals in multiple regions in the brain [area Regional observations of elevated
under the receiver operating curve (AUROC) values 0.85–
0.94 for many regions] (Supplementary Fig. 1). Tau-PET
tau-PET signal
AUROC values were improved by PVC in 30 of 47 regions Elevated tau-PET SUVr (i.e. tau-PET SUVr above the 95th
(P 4 0.05, data not shown). Among cognitively unim- percentile of young cognitively unimpaired, aged 30–49)
paired, many temporal and extra-temporal tau-PET regions was most frequently observed among cognitively unim-
also showed significant discrimination between normal and paired individuals in the amygdala region (40%), including
abnormal amyloid participants (AUROCs 0.60–0.73; 33/47 both abnormal amyloid (55%) and normal amyloid (31%)
of regions). cognitively unimpaired individuals (Fig. 2). Many other re-
gions, including extra-medial temporal and extra-temporal
regions, also had elevated tau-PET among cognitively un-
Age associations impaired. The high tau-PET SUVr values were seen outside
Among all cognitively unimpaired, tau-PET SUVr was asso- the medial temporal lobe in both temporal and extra-tem-
ciated with age in many regions. After adjusting for con- poral regions in both normal amyloid and abnormal amyl-
tinuous amyloid-PET, the positive age relationship was seen oid individuals. The percentage with elevated tau-PET was
in only the inferior temporal and amygdala regions (other significantly greater among abnormal amyloid versus
than non-specific uptake in putamen, pallidum and caud- normal amyloid cognitively unimpaired in nearly all brain
ate) (Fig. 1). Amyloid PET in this group was also associated regions. However, after adjusting for age, and correcting
with age and may be difficult to disentangle from a tau-age for multiple comparisons, this difference remained in only
association (Supplementary Fig. 2). When the cognitively seven regions (Fig. 2, asterisks). In those who had an inter-
unimpaired normal amyloid and cognitively unimpaired ab- mediate level of amyloid (1.33–1.42), two regions more
normal amyloid groups were analysed separately, most frequently had elevated tau-PET signal in the intermediate

Table 1 Characteristics of participants

Characteristic Cognitively Abnormal amyloid P-value*

unimpaired cognitively impaireda
(n = 601) (n = 86)
Age, years 0.01
Median (IQR) 68 (57, 78) 74 (64, 78)
Min, max 30, 98 52, 94
Male gender, n (%) 331 (55) 50 (58) 0.59
Education, years, median (IQR)b 16 (13, 16) 16 (12, 18) 0.21
APOE "4 positive, n (%)b 156 (27) 52 (67) 50.001
MMSE, median (IQR)b 29 (28, 29) 24 (21, 27) 50.001
Tau-PET, SUVr, median (IQR)c 1.07 (1.01, 1.13) 1.64 (1.44, 1.87) 50.001
*
Amyloid-PET, SUVr
Median (IQR) 1.34 (1.26, 1.46) 2.48 (2.15, 2.69)
41.42, n (%) 179 (30) 86 (100)

*
P-values are shown for differences between the cognitively unimpaired and cognitively impaired groups. Differences in amyloid-PET were not tested as amyloid-PET was used in
defining the groups.
a
The abnormal amyloid cognitively impaired group includes 35 (41%) individuals with amnestic mild cognitive impairment and 51 (59%) with dementia due to Alzheimer’s disease all
with amyloid-PET SUVr 4 1.42.
b
Education was missing for one cognitively unimpaired individual; MMSE was missing for four cognitively unimpaired and three cognitively impaired individuals; APOE genotype was
missing for 28 cognitively unimpaired and eight cognitively impaired individuals.
c
Tau-PET SUVr is shown for the entorhinal cortex region.
Tau imaging in ageing, Braak stage and dementia BRAIN 2018: 141; 271–287 | 275

Figure 1 Tau-PET associations with age by region and clinical group. Spearman rank correlation coefficients for the association
between regional tau-PET SUVr (without PVC) and age are shown among all cognitively unimpaired individuals in A with (light red) and without
(dark red) adjusting for amyloid-PET SUVr. Correlations of tau-PET SUVr and age are shown within normal amyloid cognitively unimpaired,
abnormal amyloid cognitively unimpaired, and abnormal amyloid cognitively impaired individuals in B–D for tau-PET values with (lighter colours)
and without (darker colours) PVC. Age was significantly correlated with tau-PET among all cognitively unimpaired in 34 of 47 regions without
adjusting for amyloid-PET SUVr and in 12 of 47 regions after adjusting for amyloid-PET. Age was significantly correlated with tau-PET without PVC
in 20/47 regions among normal amyloid cognitively unimpaired, in 10/47 regions among abnormal amyloid cognitively unimpaired, and in 38/47
regions among abnormal amyloid cognitively impaired. Tau-PETwith PVC was associated with age in most regions. Permutation tests were used to
correct P-values and confidence intervals for multiple comparisons across regions.

group than the low group (amygdala and frontal inferior unimpaired included the amygdala (Braak Stage III),
orbital) and there were no significant regional differences middle temporal pole (Braak Stage III–IV), inferior tem-
between the intermediate and high PIB groups. All cogni- poral (Braak Stage IV), hippocampus (Braak Stage III),
tively normal groups have elevated tau-PET signal in re- entorhinal cortex (Braak Stage I–II), middle temporal
gions that were widespread (temporal, frontal superior (Braak Stage IV), superior frontal orbital (Braak Stage V),
orbital, inferior occipital, frontal mid orbital, precuneus rectus (Braak Stage, uncertain), inferior occipital (Braak
and others) (Fig. 2 and Supplementary Fig. 3). Stage V), and fusiform (Braak Stage III-IV) (Fig. 2). The
entorhinal region ranked fifth in overall frequency; it was
tied for eighth in frequency among normal amyloid and
Braak staging correlation was fourth among abnormal amyloid individuals.
The 10 regions (with associated Braak Stage) ranked by the Examples of the imaging findings are shown on the right
most frequent tau-PET elevation among cognitively in Fig. 2. Tau-PET regions that showed elevated signal
276 | BRAIN 2018: 141; 271–287 V. J. Lowe et al.

Figure 2 Percentage of individuals with elevated tau-PET signal in each region with image examples. The plot on the left shows
the percentages of individuals with abnormal tau-PET signal by region within normal amyloid cognitively unimpaired (green), abnormal amyloid,
cognitively unimpaired (blue), and abnormal amyloid cognitively impaired (orange) groups. Asterisks indicate regions where the proportion of
individuals with elevated tau-PET was significantly greater among abnormal amyloid versus normal amyloid cognitively unimpaired after adjusting
for age and correcting for multiple comparisons. Abnormal amyloid cognitively impaired individuals had the higher percentage of elevated tau-PET
in all regions. Transaxial tau-PET images shown on the right show low and high SUVr ranges for normal amyloid (green arrows and circles) and
abnormal amyloid (blue arrows and circles), cognitively unimpaired participants from selected regions (arrows show the regions that correlate
with the images). Three temporal and two extra-temporal regions are shown. SUVr values (white text) are shown below each image. All images
are normalized to the same colour scale.

could appear as focal or diffuse accumulations with both (Supplementary Figs 4 and 5). The demographic character-
abnormal amyloid and normal amyloid individuals and istics of each cluster are seen in Table 2. Participants within
were seen in low and high Braak stage regions in both both the cognitively unimpaired and cognitively impaired
groups. clusters differed by age. The cognitively impaired clusters
also differed by diagnosis, gender, APOE genotype, and
MMSE. The cognitively unimpaired clusters differed by
Hierarchical clustering education, MMSE, and amyloid-PET. Box plots showing
Four cognitively unimpaired clusters and three cognitively selected regional values are shown in Fig. 3 (all regions;
impaired clusters were selected by choosing enough distinct Supplementary Fig. 6). Formal tests of differences in tau-
clusters while keeping the number of clusters tractable PET SUVr across the clusters were not done since the
Table 2 Characteristics of all participants by hierarchical clusters

Characteristic Among cognitively unimpaired individuals Among cognitively impaired individuals

Cluster 1 Cluster 2 Cluster 3 Cluster 4 P-value* Cluster 1 Cluster 2 Cluster 3 P-value*
(n = 33) (n = 163) (n = 281) (n = 124) (n = 49) (n = 31) (n = 6)
Diagnosis, n (%) * 50.001
Tau imaging in ageing, Braak stage and dementia

CU 33 (100) 163 (100) 281 (100) 124 (100) 0 (0) 0 (0) 0 (0)
aMCI 0 (0) 0 (0) 0 (0) 0 (0) 30 (61) 5 (16) 0 (0)
ADD 0 (0) 0 (0) 0 (0) 0 (0) 19 (39) 26 (84) 6 (100)
Age, years 50.001 50.001
Median (IQR) 64 (54, 76) 66 (55, 75) 67 (57, 77) 75 (67, 83) 77 (72, 81) 65 (60, 74) 55 (53, 57)
Min, max 34, 87 30, 98 30, 90 32, 94 61, 94 52, 80 52, 58
Male gender, n (%) 21 (64) 100 (61) 143 (51) 67 (54) 0.13 33 (67) 16 (52) 1 (17) 0.03
Education, years, median (IQR)a 14 (12, 16) 16 (13, 17) 16 (13, 17) 14 (12, 16) 0.004 16 (13, 18) 16 (13, 18) 14 (12, 16) 0.50
APOE "4 positive, n (%)a 9 (29) 47 (31) 71 (27) 29 (24) 0.58 25 (58) 24 (83) 3 (50) 0.05
MMSE, median (IQR)a 29 (28, 29) 29 (28, 30) 29 (28, 30) 29 (28, 29) 0.01 26 (24, 28) 22 (17, 24) 16 (13, 18) 50.001
Tau-PET, SUVr, median (IQR)b 0.92 (0.89, 0.95) 1.00 (0.96, 1.03) 1.08 (1.05, 1.11) 1.20 (1.13, 1.29) * 1.47 (1.20, 1.68) 1.82 (1.67, 2.00) 1.90 (1.84, 2.01) *
Amyloid-PET, SUVr
Median (IQR) 1.23 (1.17, 1.32) 1.31 (1.23, 1.39) 1.35 (1.27, 1.45) 1.39 (1.32, 1.64) 50.001 2.32 (2.01, 2.70) 2.55 (2.42, 2.69) 2.43 (2.35, 2.53) 0.10
41.42, n (%) 4 (12) 36 (22) 83 (30) 56 (45) 50.001 49 (100) 31 (100) 6 (100) *

*P-values are shown for overall tests of any difference in characteristic among the cognitively unimpaired clusters and any difference among the cognitively impaired clusters. Differences in diagnosis among the cognitively unimpaired clusters, tau-
PET among the cognitively unimpaired and cognitively impaired clusters, and abnormal amyloid-PET among the cognitively unimpaired clusters were not tested as the clusters were defined using these variables.
a
Education was missing for one cognitively unimpaired individual; MMSE was missing for four cognitively unimpaired and three cognitively impaired individuals; APOE genotype was missing for 28 cognitively unimpaired and eight cognitively
impaired individuals.
b
Tau-PET SUVr is shown for the entorhinal cortex region.
aMCI = amnestic MCI; CU = cognitively unimpaired.
BRAIN 2018: 141; 271–287
| 277
278 | BRAIN 2018: 141; 271–287 V. J. Lowe et al.

Figure 3 Box plots of regional tau-PET SUVr by hierarchical clusters. Box plots of tau-PET SUVr for nine regions of interest by the
four clusters among cognitively unimpaired individuals and three clusters among cognitively impaired individuals. Clinical diagnosis and amyloid-
PET status is represented by different colours: normal amyloid cognitively unimpaired, green; abnormal amyloid cognitively unimpaired, blue;
abnormal amyloid amnestic MCI (aMCI), purple; abnormal amyloid ADD, orange. The cognitively unimpaired groups have a gradually increasing
tau-PET signal in all cortical regions. Greater tau-PET signal is seen in temporal and extra-temporal cortical regions (frontal, posterior cingulate,
parietal, angular and temporal inferior) in the cognitively impaired cluster 3 compared to cluster 2 but similar tau-PET signal in the entorhinal
cortex was seen in these two groups. Those in the cognitively impaired cluster 1 are more similar to the cognitively unimpaired cluster 4 than the
other cognitively impaired clusters. Formal tests of differences in tau-PET among the clusters were not performed since the clusters were defined
by tau-PET.

clusters were defined by tau-PET, but visual inspection of and 39% ADD participants. In general, there was a gradi-
the regional tau-PET values show some interesting topo- ent in tau-PET signal across the three cognitively impaired
graphical distributions of tau-PET among the clusters. In clusters with 1 5 2 5 3. Cluster 3 was the smallest cluster,
the cognitively unimpaired clusters 1 through 4 there was younger, all were ADD, and had visually different topo-
increasing tau-PET signal in all regions. The cognitively graphical distributions of tau-PET signal. In particular,
unimpaired cluster 4 had similar tau-PET SUVr levels in cluster 3 had higher tau-PET SUVr in many regions com-
some regions (frontal, inferior parietal, and posterior cin- pared to the other cognitively impaired groups with ex-
gulate regions for example) to the cognitively impaired amples shown in Fig. 3 as mid-frontal, superior frontal,
cluster 1 group in which there were 61% amnestic MCI posterior cingulate, inferior parietal, angular, and inferior
Tau imaging in ageing, Braak stage and dementia BRAIN 2018: 141; 271–287 | 279

temporal regions (AUROC 0.89–1.0). This was in contrast increasing from cognitively unimpaired normal amyloid to
to other regions such as the medial temporal regions cognitively unimpaired abnormal amyloid and from amnestic
(amygdala, entorhinal cortex and hippocampus shown) MCI to ADD individuals. Differences in tau-PET SUVr were
where the SUVr were more similar between cluster 3 and seen among the three cognitively unimpaired groups and
cluster 2 (AUROC 0.54–0.65). among the four cognitively impaired groups for all nine re-
gions shown in Fig. 4 (P 5 0.01 after adjusting for multiple
Clinical characteristic comparisons). Variation in tau-PET signal across cognitively
impaired groups was comparable to hierarchical clustering
subcategorization with tau-PET signal being appreciably higher in extra-tem-
A priori grouping of participants by clinical diagnosis, age, poral regions for younger versus older-age cognitively
and amyloid-PET abnormality showed tau-PET signal impaired individuals (Fig. 4 and Supplementary Fig. 7).

Figure 4 Box plots of regional tau-PET SUVr by age, amyloid-PET status and diagnosis subcategorization. Box plots of tau-PET
SUVr for nine regions of interest by clinical diagnosis, amyloid-PET status, and age. Clinical diagnosis and amyloid-PET status are represented with
different colours: normal amyloid cognitively unimpaired, green; abnormal amyloid cognitively unimpaired, blue; abnormal amyloid amnestic MCI
(aMCI), purple; abnormal amyloid ADD, orange. Tau-PET SUVr differed significantly (P 5 0.01), after adjusting for multiple comparisons, among
the three cognitively unimpaired groups and among the four cognitively impaired groups for all nine regions shown in the figure. There is greater
tau signal in the medial and extra medial temporal regions (inferior temporal for example; other regions seen on Supplementary Fig. 7) than in the
cognitively unimpaired 450 normal amyloid group versus the cognitively unimpaired normal amyloid 550 group. The box plots also show a
similar pattern of increased tau-PET signal in younger age ADD as compared to the cluster analysis with much higher extra-temporal tau-PET
signal. However, there are participants in the younger age ADD group with tau-PET signal similar to the older age ADD group.
280 | BRAIN 2018: 141; 271–287 V. J. Lowe et al.

In contrast, medial temporal tau-PET signal (entorhinal and signal compared to cognitively impaired cluster 1 in the
hippocampal regions for example) was more similar be- temporal, parietal and posterior cingulate regions.
tween younger and older cognitively impaired individuals The voxel-wise image data for the clinical-characteristic
(560 versus 570 ADD) in the clinical subcharacterization. groups are displayed as contrasts to show significant differ-
The amnestic MCI group generally had lower tau-PET ences in tau-PET signal between the groups (Fig. 6). In
signal than the ADD groups; however, there were a normal amyloid, cognitively unimpaired older participants,
number of amnestic MCI individuals with very high tau- greater medial temporal, extra-medial temporal and basal
PET SUVr in some regions (angular and parietal regions, frontal tau-PET signal was seen than in younger normal
for example). The cognitively unimpaired groups showed amyloid, cognitively unimpaired participants. Amnestic
less variation in magnitude of tau-PET signal across regions MCI participants had greater temporal, parietal, posterior
than what was seen in the hierarchical clustering. In the cingulate and frontal tau-PET signal than cognitively unim-
hierarchical clustering, several cognitively unimpaired paired participants. Similar to the clustering findings, in the
normal amyloid participants are clustered with cognitively clinically subcategorized group contrasts, younger ADD
unimpaired abnormal amyloid individuals (e.g. green dots show greater contrast of tau-PET signal and more frontal
in the cognitively unimpaired cluster 4 in the posterior cin- tau-PET signal with medial temporal sparing than older
gulate, parietal and angular regions, Fig. 3) and are also ADD.
similar in tau-PET SUVr to cluster 1 cognitively impaired
individuals. The hierarchical clustering may be better at Tau patterns and relationship with
identifying cognitively unimpaired individuals with possible
preclinical Alzheimer’s disease tau-PET signal patterns (e.g.
clinical phenotypes
cognitively unimpaired cluster 4 individuals) than simply The clinical phenotypes of the cognitively impaired clusters
grouping cognitively-unimpaired individuals by age and differed. In the cognitively impaired cluster 3, 5/6 had the
amyloid abnormality. rare dysexecutive phenotype of ADD while 1/6 was
described as younger-age onset, memory-predominant
ADD. The SPM analysis and region of interest analyses
Voxel-wise analyses showed more tau-PET signal in the frontal regions in the
cognitively impaired cluster 3. Interestingly, in the cogni-
The voxel-wise image data are displayed as summed image
tively impaired cluster 2, 3/31 were called dysexecutive
data from each individual cluster for a descriptive display
ADD and another one had a mild mixed memory and
(Supplementary Fig. 8) and as contrasts between the clus-
dysexecutive presentation. Other individuals in cluster 2
ters to descriptively show tau-PET signal differences (Fig.
were either MCI or ADD, of which additional clinical find-
5). As the clustering used tau-PET SUVr to identify groups
ings of sleep apnoea (n = 1), topographical agnosia (n = 1),
with similar tau patterns, the reported T-statistic in the
and limbic ADD (n = 1) were seen. The dysexecutive ADD
colour bar for the contrasts in these clusters is descriptive.
individuals in cognitively impaired cluster 2 had less frontal
Cognitively unimpaired clusters topographically look simi-
tau signal than those in cluster 3 on visual inspection (data
lar with greater widespread brain tau-PET signal progress-
not shown), consistent with the clustering results. The cog-
ing from cognitively unimpaired cluster 1 to cognitively
nitively impaired cluster 3 participants had the lowest
unimpaired cluster 4 on the summed data. Greater diffuse
MMSE.
temporal and parietal tau-PET signal is visible on the
summed images in cognitively unimpaired cluster 4 or cog-
nitively unimpaired cluster 3 and is seen on the SPM con-
trast as well. On SPM analyses, more contrast between
Discussion
cognitively unimpaired cluster 2 and cognitively unimpaired Determining the distribution of neurofibrillary tangles
cluster 1 is seen in the frontal regions and continues to be a across the lifespan is important to better elucidate the se-
prominent finding in the other cognitively unimpaired clus- quential evolution of the pathophysiology of ADD. In this
ter contrasts. Differences are seen in many other regions of study we used tau-PET to infer the regional characteristics
the brain between the cognitively unimpaired clusters. of neurofibrillary tangles seen in cognitively unimpaired
Cognitively impaired cluster 3 (mostly younger-age onset and cognitively impaired individuals and correlated topo-
ADD) had the highest levels of tau-PET across most regions graphical tau-PET patterns with clinical diagnosis, age, and
on the region of interest analysis and has greater tau-PET amyloid status. We emphasize four major findings. First,
signal contrast in the frontal, temporal, parietal, posterior tau-PET SUVr was modestly associated with age through-
cingulate and angular regions as compared to other cogni- out most regions of the brain in cognitively unimpaired
tively impaired clusters. Interestingly, there was relative individuals. Second, this widespread elevated tau-PET
sparing of incremental tau-PET signal in medial temporal signal in the brain was seen in both normal amyloid and
regions in cognitively impaired cluster 3 as compared to abnormal amyloid status, cognitively unimpaired individ-
cognitively impaired cluster 2 or cognitively impaired clus- uals (i.e. signal was not confined to medial temporal re-
ter 1. Cognitively impaired cluster 2 has higher tau-PET gions). Third, the distribution of tau-PET signal in
Tau imaging in ageing, Braak stage and dementia BRAIN 2018: 141; 271–287 | 281

Figure 5 Comparison of the differences in tau-PET signal between hierarchical clusters. Voxel-wise tau-PET findings that are greater
between pairs of hierarchical clusters are shown. The upper three rows show the comparisons of cognitively unimpaired clusters (green text).
The lower four rows show cognitively impaired cluster comparisons (red text). There is greater tau-PET signal diffusely in cognitively unimpaired
2 versus cognitively unimpaired 1 clusters. Greater tau-PET signal is seen in cognitively unimpaired 3 and cognitively unimpaired 4 versus
cognitively unimpaired 1 in the whole temporal and parietal lobes (arrows). Frontal accumulation is incrementally greater also. The incremental
tau-PET signal seen between the different cognitively unimpaired clusters involves many regions of the brain outside of the temporal lobe. The
magnitude of the increased tau-PET signal in the cognitively impaired clusters is greatest in the frontal lobes and is similar to the findings in Fig. 3.
The cognitively impaired cluster 1 shows increased inferior temporal, medial temporal and posterior cingulate tau-PET signal (arrows) as
compared to cognitively unimpaired cluster 4. Greater tau-PET signal in cognitively impaired 2 versus cognitively impaired 1 is seen in the
temporal, parietal and posterior cingulate regions and less so in the frontal lobe (arrow). Cognitively impaired cluster 3 showed greater tau-PET
signal in the frontal, temporal, parietal, posterior cingulate and angular regions than cognitively impaired 1 or cognitively impaired 2 clusters with
pronounced increase in the frontal lobe versus cognitively impaired cluster 1 (arrows). All results were family-wise error (FWE) corrected,
P 5 0.05 and the colour bar shows the T-statistic range for all contrasts but is descriptive as the clustering was defined by using tau-PET SUVr.
282 | BRAIN 2018: 141; 271–287 V. J. Lowe et al.

Figure 6 Comparison of the differences in tau-PET signal between clinically characterized groups. Voxel-wise tau-PET findings that
are greater between pairs of the clinically characterized groups are shown. The upper two rows show the comparisons of cognitively unimpaired
groups by age and amyloid status (green text). The lower five rows show comparisons of cognitively impaired groups by diagnosis comparisons
(red text). There is greater tau-PET signal in the inferior temporal lobes (black arrow) (in additional to medial temporal lobes) and frontal orbital
regions (dashed arrow, shown on inferior view) in the cognitively unimpaired 450 normal amyloid group versus the cognitively unimpaired
normal amyloid 550 group. Greater tau-PET signal is seen in the temporal, parietal and the frontal lobes (dashed arrow) in the cognitively
unimpaired abnormal amyloid group versus the cognitively unimpaired normal amyloid 550 group. The amnestic MCI group shows increased
temporal, parietal, posterior cingulate, and frontal tau-PET signal as compared to the cognitively unimpaired abnormal amyloid group and a similar
pattern is also seen in ADD 570 versus amnestic MCI. This pattern is seen in ADD 60–69 versus ADD 570 but is more diffuse. When the
ADD 560 is contrasted to the other ADD groups, incremental frontal tau-PET signal is prominent (dashed arrow) but medial temporal
differences are relatively small (arrow). All results were few corrected, P 5 0.05 and the colour bar shows the T-statistic range for all contrasts.
A = amyloid; aMCI = amnestic MCI; CU = cognitively unimpaired.
Tau imaging in ageing, Braak stage and dementia BRAIN 2018: 141; 271–287 | 283

cognitively unimpaired individuals suggested a pattern of tau-PET age associations were limited to inferior temporal
early neurofibrillary tangle deposition with similarities to and amygdala regions. Interestingly, in the combined cog-
Braak neurofibrillary tangle staging but also with nitively unimpaired group, some regions were also nega-
exceptions. And fourth, regional hierarchical clustering of tively correlated with age after adjusting for amyloid-PET
tau-PET signal and clinical characteristic-based grouping of SUVr. This may be due to an over-correction or be an
individuals revealed variable patterns of neurofibrillary artefact of the analysis, or could be explained if the older
tangle topographical distributions within cognitively unim- abnormal amyloid individuals were ‘resilient’ to amyloid
paired and cognitively impaired subgroups. These four ob- (i.e. remained cognitively unimpaired rather than progress-
servations demonstrate common principles of the ing to dementia) and had less tau than younger abnormal
pathophysiology of ADD: namely that neurofibrillary amyloid individuals (Fig. 1A). In support of this idea, when
tangle abundance is in part an ageing phenomenon; that abnormal amyloid and normal amyloid groups were ana-
b-amyloidosis may be a trigger to neurofibrillary tangle lysed separately, positive age correlation was seen in most
progression but widespread, low level tangle deposition regions. Extra-temporal tau-PET signal in normal amyloid,
can occur in the absence of amyloidosis, and that regional cognitively unimpaired older adults—similar to our find-
expansion of increasing neurofibrillary tangle burden ings—may have been subtly present in prior work, but
begins during the preclinical phase of the disease and that was not specifically discussed [see Fig. 1 in Scholl et al.,
neurofibrillary tangle topographical distribution in the (2016), with older normal amyloid individuals showing dif-
brain can have clinical relevance in impaired individuals. fuse tau-PET signal greater than the younger group].
With our large sample size, we were able to detect a Interestingly, in our cognitively impaired group, higher
significant, but modest age relationship with tau-PET tau SUVr was associated with younger age in most regions.
signal in many regions of the brain among (i.e. not just This could be explained by younger-onset probable ADD
medial temporal) cognitively unimpaired individuals. In a (cognitively impaired cluster 3) having more severe tau
prior tau-PET study, the association between increasing age burden versus older-onset participants and is consistent
and higher tau-PET among 38 cognitively unimpaired par- with our cluster findings (Table 2 and Fig. 5).
ticipants was limited to the medial temporal structures, ven- We evaluated the percentage of individuals with elevated
tral frontal cortex and insula (Scholl et al., 2016). In other tau-PET signal in each region among cognitively unim-
studies, age was not associated with tau-PET signal (Brier paired individuals with normal and abnormal amyloid
et al., 2016; Schwarz et al., 2016). In one recent study, tau- status and found that elevated tau-PET signal was seen
PET signal was not seen outside of the medial temporal frequently in medial temporal lobe regions but also in
lobe in normal amyloid, cognitively unimpaired older many extra-temporal brain regions among those with
adults (n = 58, 450 years) as compared to younger adults either normal or abnormal amyloid status (Fig. 2). This
(Pontecorvo et al., 2017). In contrast, our tau-PET findings contrasts with previous reports where participants with
in a larger cognitively unimpaired group, showed age-asso- normal amyloid were said to have no tau-PET signal ele-
ciated tau-PET signal in most regions of the brain— vation outside of the medial temporal lobe (Scholl et al.,
whether individuals had normal or abnormal amyloid 2016). Most regional elevations showed a trend of being
status. The observed broader brain distribution of more frequent in cognitively unimpaired participants with
age-related, tau-PET signal in cognitively unimpaired par- abnormal amyloid, with some (n = 7) still significantly more
ticipants seen in our group may differ from prior reports frequent than those with normal amyloid after correction
because of the size of our population, our recruitment for multiple comparisons and age. These regional tau-PET
methods (the cognitively unimpaired group is from a popu- findings have similarities to Braak descriptions of early
lation-based sample), or image analysis methodological dif- neurofibrillary tangle extent, but are not identical, with
ferences. Importantly, for image analyses, we defined some elevated tau-PET regions being located in Braak
abnormal tau-PET signal as being greater than that seen Stages III–VI areas. This was true for tau-PET signal in
in a group of young cognitively unimpaired individuals both normal and abnormal amyloid, cognitively unim-
(550 years of age) and the definitions of abnormality are paired participants and would suggest that high Braak
specific for each region rather than for a meta-region of Stage (extra-medial temporal and extra-temporal neocor-
interest or a large brain region. This has some contrasts tex) deposition can occur independent of amyloid. These
to prior reports that used multi-region-composite regions, findings could be concordant with primary age-related
often designed to mimic Braak stage regional anatomy tauopathy (PART) (Crary et al., 2014) and with our own
(Johnson et al., 2016; Scholl et al., 2016; Schwarz et al., pre-tau PET observations in persons with a suspected non-
2016). Alzheimer pathophysiology (SNAP) designation (Knopman
Regional tau-PET signal in the cognitively unimpaired et al., 2013). These findings are supported by prior autopsy
group was shown with and without adjustment for con- and tissue data. In Braak’s 1996 publication (Braak et al.,
tinuous amyloid-PET SUVr. It is difficult to disentangle 1996), amyloid deposition did not always precede neuro-
the effects of amyloid on tau-PET signal as amyloid also fibrillary tangle deposition. Importantly, the neurofibrillary
has an age relationship in the population (Jack et al., tangles in PART is 3R/4R and should be identified by
2014). After adjustment for amyloid-PET, the positive tau-PET. We have previously demonstrated binding of
284 | BRAIN 2018: 141; 271–287 V. J. Lowe et al.

AV-1451 to PART tau by autoradiography as compared to unimpaired versus ADD at the time of autopsy, consistent
immunohistochemistry (Lowe et al., 2016). Together, these with the findings on tau-PET of an increasing presence of
data provide biomarker evidence to suggest that PART, neurofibrillary tangles throughout the brain with ageing.
commonly thought to be mostly restricted to the medial The biological significance of widespread tau-PET signal
temporal lobe and other allocortical areas, or some other in abnormal amyloid and normal amyloid cognitively un-
cause of tau-PET signal elevation without amyloid, is a impaired individuals is unknown. The medial temporal ele-
relatively common finding in cognitively unimpaired and vations seen in cognitively unimpaired cluster 4 could
could be more widespread in the brain than previously represent a classic appearance of pathology that could be
thought. Its clinical significance, however, remains uncer- expected to lead to ADD. Further longitudinal imaging and
tain and correlation with the development of cognitive im- follow-up will be needed to determine with any certainty
pairment, longitudinal data and clinicopathological the significance of these findings. Nevertheless, these
verification will be needed to further understand its import- tau-PET findings provide observations that could add im-
ance. Indeed we recently found poorer cognitive perform- portant insight to the biology of neurofibrillary tangle de-
ance and atrophy of the head of the left hippocampus in velopment in Alzheimer’s disease and direct subsequent
participants with PART suggesting PART may not be a studies. Some current thinking suggests a topographical
silent pathology (Josephs et al., 2017). progression of neurofibrillary tangle, possibly facilitated
Two recent studies reported increased tau-PET only in by trans-synaptic spread of misfolded tau protein (Guo
medial temporal or temporal structures in groups of 15 and Lee, 2011). Trans-neural spread through network con-
and 56 cognitively unimpaired individuals, respectively nections could facilitate this propagation to distant areas
(Johnson et al., 2016; Ossenkoppele et al., 2016). within a functional network (Stancu et al., 2015). The tau-
Furthermore, extra-temporal neurofibrillary tangle findings PET data presented herein support the development of
are not highlighted in some large ageing population aut- widespread, age-related neurofibrillary tangles in the
opsy studies (Braak and Braak, 1997). These differences brain. This could be a manifestation of trans-neuronal
from our findings could be partially explained by the spread but at an age years before symptomatic disease.
tissue sampling limitations inherent in autopsy studies and Worsening of neurofibrillary tangle pathology, associated
possibly the smaller sample sizes in previous tau-PET with a trigger, possibly amyloid and/or neural network-
studies. Our tau-PET findings suggest that high Braak related activity, could potentially lead to progressive neuro-
neurofibrillary tangle stages may be frequently detected fibrillary tangle pathology that eventually causes clinical
by tau-PET in cognitively unimpaired subjects. Supportive symptoms. Additional recent observations show that sol-
of these PET findings are smaller cohort autopsy data uble phosphorylated high molecular weight tau, found in
showing that: (i) advanced neurofibrillary tangle stages the extracellular space, can trigger tau propagation, and
(IV–VI) are seen in cognitively unimpaired and demented provide supporting biochemical data of another hypothesis
individuals with considerable overlap (Gertz et al., 1996); of how widespread tau development throughout the brain
(ii) widespread isocortical neurofibrillary tangles were ‘less could occur (Takeda et al., 2015). These ideas of tau
frequent’ but present in cognitively unimpaired (Arriagada spread to distant, non-adjacent brain regions may be con-
et al., 1992); (iii) ‘a few scattered’ neurofibrillary tangles sistent with our findings. Widespread tau propagation
are clearly seen in the isocortex in cognitively unimpaired could provide the milieu for further neurofibrillary tangle
(Braak and Braak, 1991); and (iv) only small numbers of development, again with the aid of other triggers like amyl-
normal ageing and ADD patients conform in all respects to oid and/or neural network-related activity, in more specific,
Braak neurofibrillary tangle hierarchy (Gertz et al., 1998). disease-related regions. Our data show that age-related,
In addition, autopsy data also support progressive increase widespread elevated tau-PET signal becomes more frequent
in neurofibrillary tangle frequency with age in limited re- and of greater magnitude when amyloid is elevated and
gions (Braak et al., 1996). In concordance with earlier aut- supports these ideas. Longitudinal and clinical–pathological
opsy data, the magnitude of tau-PET SUVr values among validation studies are needed to evaluate these hypotheses.
cognitively unimpaired was relatively low compared to cog- Findings from the clustering analyses and subcategoriza-
nitively impaired, suggesting that modest neurofibrillary tion by clinical findings provided several important insights.
tangle involvement as suggested by tau-PET is a reasonable The clustering analyses defined three cognitively impaired
correlate to autopsy data (Arriagada et al., 1992). groups and demonstrated that younger-age onset ADD had
Importantly, quantitative analysis of neurofibrillary tangles more diffusely distributed and higher tau-PET SUVr signal
has shown neurofibrillary tangle density/mm2 in the super- with some cortical region preferences (e.g. frontal regions,
ior frontal cortex to be 4.9  0.5 versus 7.7  3.3 in cog- posterior cingulate, and parietal regions being greater) as
nitively unimpaired versus ADD (a relatively modest compared to older-age onset ADD. The cognitively
difference), and in layer II of the entorhinal cortex to be impaired clusters differed by age, diagnosis, gender, and
7.4  0.8 versus 59.1  12.4 in cognitively unimpaired APOE genotype status suggesting clinically relevant impli-
versus ADD (Bouras et al., 1994). Together, these autopsy cations of the tau-PET patterns. Medial temporal regions
data support fewer, but not absent, neurofibrillary tangles were also relatively spared in the younger-age onset
in extra-temporal and temporal regions in cognitively cognitively impaired cluster relative to the other cognitively
Tau imaging in ageing, Braak stage and dementia BRAIN 2018: 141; 271–287 | 285

impaired clusters. In the cognitively impaired cluster 3, 5/6 addition of widespread brain tau-PET signal also being
had a dysexecutive phenotype of ADD while 1/6 was able to discriminate abnormal amyloid and normal amyl-
described as younger-age onset ADD. Behavioural or dys- oid, cognitively unimpaired groups.
executive ADD is a rare ADD phenotype in which clinical- Limitations include tau-PET signal in off-target sites. We
anatomic correlations are poorly understood. Previous have previously shown that tau-PET signal can be seen in
work has shown that this group has some biomarker dif- off-target sites and other non-tauopathies (Lowe et al.,
ferences (atrophy on MRI) as compared to behavioural 2016). For example, retention in the choroid plexus could
variant frontotemporal dementia (Ossenkoppele et al., bias medial temporal lobe regional findings. We think that
2015). Differences in glucose metabolism and amyloid dis- this influence is likely minimal as we found that only four
tribution in younger onset ADD as compared to older onset cases in the cognitively unimpaired group had elevated tau
ADD have also been described in prior studies that showed signal only in the entorhinal cortex of the medial temporal
more parietal abnormalities in younger onset ADD lobe regions and only 1/4 had abnormal entorhinal cortex
(Ossenkoppele et al., 2012, 2015). The clustering method tau-PET signal as the sole finding. Two of these four were
we used grouped mostly dysexecutive ADD phenotype par- unlikely to have a choroid plexus bias because low choroid
ticipants in a single cluster while the clinically based sub- plexus retention was seen visually. This demonstrates that
categorization included five other individuals in the choroid plexus ‘bleed in’ is likely of limited importance in
ADD 560 year-old group. These five individuals in the group-wise analyses. We also showed that entorhinal
clinically based subcategorization did not have highly ele- cortex tau-PET signal has significant clinical relevance in
vated frontal tau-PET signal (two of whom had a dysexe- region of interest analysis (Fig. 4) supporting the idea of
cutive phenotype) as seen in the cognitively impaired cluster limited bias. We also selected a group of participants
3 (on visual review) and were included in other clusters by aged 550 to determine tau-PET ‘normality.’ Early tau de-
the clustering method. While the numbers are too small to position may be present in some younger participants. In
be definitive, the finding suggests that topographical tau-
any case, this would likely conservatively bias our results.
PET distribution may be helpful in younger onset ADD
Autopsy data correlated with ante-mortem imaging and
phenotype characterization and that different regional pat-
longitudinal comparisons will be needed for verification
terns of tau-PET signal are associated with different ADD
of the implications of age-related or widely disseminated
phenotypes. Future work in non-amnestic ADD patients
increases in tau-PET signal and validate cut-points for
and in persons with other non-Alzheimer’s disease-tauopa-
tau-PET. Lastly, our analysis averaged region of interest
thies will be important to assess the variability of tau
values from different sides of the brain. In some instances
topographical distributions that may be present in the
asymmetry in tau-PET signal between sides may exist (Fig.
tau-disease spectrum.
2) and could be relevant to individual clinical presentations.
We hypothesized that the clustering analysis would detect
Further analysis using side-specific tau-PET data and clin-
groups of individuals with distinct regional patterns of tau-
ical correlation will be needed to assess this observation.
PET signal and particular phenotypes in the cognitively
unimpaired group similar to how it performed in the cog-
nitively impaired group. Increasing tau-PET signal was seen
globally in cognitively unimpaired clusters 1 through 4 and
we saw greater temporal and parietal tau-PET signal in the
Acknowledgements
highest tau-PET signal cognitively unimpaired clusters. We We would like to greatly thank AVID Radiopharmaceuti-
may not have been able to detect additional patterns in cals, Inc., for their support in supplying the AV-1451 pre-
cognitively unimpaired individuals possibly because our cursor, chemistry production advice and oversight, and
cognitively unimpaired cluster groups were large and not FDA regulatory cross-filing permission and documentation
granular enough to detect early development of specific needed for this work.
ADD or atypical ADD phenotypes. It will be interesting
to see in the future if defining tau-PET topographically
associated groups by other clustering selections or other
clinical methods will be better at identifying tau patterns
that can predict ADD phenotypes in preclinical stages.
Funding
Lastly, highly accurate group-wise separation of cogni- This research was supported by NIH grants, P50
tively unimpaired and cognitively impaired groups with AG016574, R01 NS89757, R01 NS089544, R01
tau-PET was seen using almost any cortical region of the DC10367, U01 AG006786, R21 NS094489, by the
brain. Similar findings by others have been seen in limited Robert Wood Johnson Foundation, The Elsie and Marvin
regions (Johnson et al., 2016; Ossenkoppele et al., 2016; Dekelboum Family Foundation, The Liston Family
Schwarz et al., 2016). While medial temporal regions had Foundation and by the Robert H. and Clarice Smith and
some of the highest tau-PET SUVr and AUROC values, Abigail van Buren Alzheimer’s Disease Research Program,
high discriminative ability was seen throughout most The GHR Foundation, Foundation Dr Corinne Schuler and
brain regions. Our data support prior findings with the the Mayo Foundation.
286 | BRAIN 2018: 141; 271–287 V. J. Lowe et al.

Bouras C, Hof PR, Giannakopoulos P, Michel JP, Morrison JH.

Conflicts of interest Regional distribution of neurofibrillary tangles and senile plaques
in the cerebral cortex of elderly patients: a quantitative evaluation
V.J.L consults for Bayer Schering Pharma, Piramal Life of a one-year autopsy population from a geriatric hospital. Cereb
Sciences and Merck Research and receives research support Cortex 1994; 4: 138–50.
from GE Healthcare, Siemens Molecular Imaging, AVID Braak H, Braak E. Neuropathological stageing of Alzheimer-related
changes. Acta Neuropathol 1991; 82: 239–59.
Radiopharmaceuticals and the NIH (NIA, NCI). P.F. re-
Braak H, Braak E. Frequency of stages of Alzheimer-related lesions in
ceives research support from GE Healthcare, Siemens different age categories. Neurobiol Aging 1997; 18: 351–7.
Molecular Imaging, AVID. B.F.B receives royalties from Braak H, Braak E, Bohl J, Reintjes R. Age, neurofibrillary changes, A
the publication of Behavioral Neurology of Dementia, beta-amyloid and the onset of Alzheimer’s disease. Neurosci Lett
serves on the Scientific Advisory Board for the Tau 1996; 210: 87–90.
Brier MR, Gordon B, Friedrichsen K, McCarthy J, Stern A,
Consortium, and receives research support from GE
Christensen J, et al. Tau and Abeta imaging, CSF measures, and
Healthcare, Axovant Sciences, the NIH, and the cognition in Alzheimer’s disease. Sci Transl Med 2016; 8: 338ra66.
Mangurian Foundation. K.A.J. receives research support Clark CM, Pontecorvo MJ, Beach TG, Bedell BJ, Coleman RE,
from the NIH (NIDCD NINDS and NIA). K.K. receives Doraiswamy PM, et al. Cerebral PET with florbetapir compared
research grants from the NIH/NIA. M.M.Ma. receives re- with neuropathology at autopsy for detection of neuritic amyloid-
beta plaques: a prospective cohort study. Lancet Neurol 2012; 11:
search support from the NIH (NIA and NIDCD). M.M.Mi
669–78.
receives research support from the NIH/NIA, Department Crary JF, Trojanowski JQ, Schneider JA, Abisambra JF, Abner EL,
of Defense, and Biogen and consults for Lysosomal Alafuzoff I, et al. Primary Age-related Tauopathy (PART): a
Therapeutics, Inc., and Eli Lilly & Company. R.O.R. re- common pathology associated with human aging. Acta
ceives research support from the NIH and a research grant Neuropathol 2014; 128: 755–66.
from F. Hoffman-La Roche. D.S.K. serves on a Data Safety Driscoll I, Troncoso JC, Rudow G, Sojkova J, Pletnikova O, Zhou Y,
et al. Correspondence between in vivo (11)C-PiB-PET amyloid ima-
Monitoring Board for Lundbeck Pharmaceuticals and for ging and postmortem, region-matched assessment of plaques. Acta
the DIAN study; is an investigator in clinical trials spon- Neuropathol 2012; 124: 823–31.
sored by TauRX Pharmaceuticals, Lilly Pharmaceuticals, Duyckaerts C, Bennecib M, Grignon Y, Uchihara T, He Y, Piette F,
Biogen and the Alzheimer’s Disease Cooperative Study; et al. Modeling the relation between neurofibrillary tangles and in-
and receives research support from the NIH. R.C.P. is a tellectual status. Neurobiol Aging 1997; 18: 267–73.
Fawaz MV, Brooks AF, Rodnick ME, Carpenter GM, Shao X,
consultant for Biogen, Roche, Inc., Merck, Inc. and
Desmond TJ, et al. High affinity radiopharmaceuticals based upon
Genentech, Inc.; receives publishing royalties from Mild lansoprazole for PET imaging of aggregated tau in Alzheimer’s dis-
Cognitive Impairment (Oxford University Press, 2003), ease and progressive supranuclear palsy: synthesis, preclinical evalu-
and receives research support from the National Institute ation, and lead selection. ACS Chem Neurosci 2014; 5: 718–30.
of Health. C.R.J. receives research support from the NIH/ Gertz HJ, Xuereb J, Huppert F, Brayne C, McGee MA, Paykel E, et al.
Examination of the validity of the hierarchical model of neuropatho-
NIA, and the Alexander Family Alzheimer’s Disease
logical staging in normal aging and Alzheimer’s disease. Acta
Research Professorship of the Mayo Foundation. Neuropathol 1998; 95: 154–8.
H.J.W., M.L.S., S.D.W., T.M.T., M.K.P., M.E.M., Gertz HJ, Xuereb JH, Huppert FA, Brayne C, Kruger H, McGee MA,
D.T.J., P.V., J.G-R. and C.G.S. report no disclosures. et al. The relationship between clinical dementia and neuropatho-
logical staging (Braak) in a very elderly community sample. Eur
Arch Psychiatry Clin Neurosci 1996; 246: 132–6.
Guo JL, Lee VM. Seeding of normal Tau by pathological Tau confor-
Supplementary material mers drives pathogenesis of Alzheimer-like tangles. J Biol Chem
2011; 286: 15317–31.
Supplementary material is available at Brain online. Hashimoto H, Kawamura K, Igarashi N, Takei M, Fujishiro T, Aihara
Y, et al. Radiosynthesis, photoisomerization, biodistribution, and
metabolite analysis of 11C-PBB3 as a clinically useful PET probe
References for imaging of tau pathology. J Nucl Med 2014; 55: 1532–8.
Jack CR Jr, Wiste HJ, Weigand SD, Rocca WA, Knopman DS, Mielke
American Psychiatric Association. Diagnostic and statistical manual of MM, et al. Age-specific population frequencies of cerebral beta-
mental disorders. 4th edn., text revision. Washington, DC: American amyloidosis and neurodegeneration among people with normal cog-
Psychiatric Association; 2000. nitive function aged 50-89 years: a cross-sectional study. Lancet
Arriagada PV, Marzloff K, Hyman BT. Distribution of Alzheimer-type Neurol 2014; 13: 997–1005.
pathologic changes in nondemented elderly individuals matches the Jack CR Jr, Wiste HJ, Weigand SD, Therneau TM, Lowe VJ,
pattern in Alzheimer’s disease. Neurology 1992; 42: 1681–8. Knopman DS, et al. Defining imaging biomarker cut points for
Ashburner J, Friston KJ. Unified segmentation. Neuroimage 2005; 26: brain aging and Alzheimer’s disease. Alzheimers Dement 2017; 13:
839–51. 205–16.
Avants BB, Epstein CL, Grossman M, Gee JC. Symmetric diffeo- Johnson KA, Gregas M, Becker JA, Kinnecom C, Salat DH, Moran
morphic image registration with cross-correlation: evaluating auto- EK, et al. Imaging of amyloid burden and distribution in cerebral
mated labeling of elderly and neurodegenerative brain. Med Image amyloid angiopathy. Ann Neurol 2007; 62: 229–34.
Anal 2008; 12: 26–41. Johnson KA, Schultz A, Betensky RA, Becker JA, Sepulcre J, Rentz D,
Bennett DA, Schneider JA, Wilson RS, Bienias JL, Arnold SE. et al. Tau positron emission tomographic imaging in aging and early
Neurofibrillary tangles mediate the association of amyloid load Alzheimer disease. Ann Neurol 2016; 79: 110–19.
with clinical Alzheimer disease and level of cognitive function. Josephs KA, Murray ME, Tosakulwong N, Whitwell JL, Knopman
Arch Neurol 2004; 61: 378–84. DS, Machulda MM, et al. Tau aggregation influences cognition
Tau imaging in ageing, Braak stage and dementia BRAIN 2018: 141; 271–287 | 287

and hippocampal atrophy in the absence of beta-amyloid: a clinico- Roberts RO, Geda YE, Knopman DS, Cha RH, Pankratz VS, Boeve
imaging-pathological study of primary age-related tauopathy BF, et al. The Mayo Clinic Study of Aging: design and sampling,
(PART). Acta Neuropathol 2017; 133: 705–15. participation, baseline measures and sample characteristics.
Kantarci K, Yang C, Schneider JA, Senjem ML, Reyes DA, Lowe VJ, Neuroepidemiology 2008; 30: 58–69.
et al. Antemortem amyloid imaging and beta-amyloid pathology in a Sabbagh MN, Cooper K, DeLange J, Stoehr JD, Thind K, Lahti T,
case with dementia with Lewy bodies. Neurobiol Aging 2012; 33: et al. Functional, global and cognitive decline correlates to accumu-
878–85. lation of Alzheimer’s pathology in MCI and AD. Curr Alzheimer
Klunk WE, Engler H, Nordberg A, Wang Y, Blomqvist G, Holt DP, Res 2010; 7: 280–6.
et al. Imaging brain amyloid in Alzheimer’s disease with Pittsburgh Scholl M, Lockhart SN, Schonhaut DR, O’Neil JP, Janabi M,
Compound-B. Ann Neurol 2004; 55: 306–19. Ossenkoppele R, et al. PET imaging of Tau deposition in the
Knopman DS, Jack CR Jr, Wiste HJ, Weigand SD, Vemuri P, Lowe aging human brain. Neuron 2016; 89: 971–82.
VJ, et al. Brain injury biomarkers are not dependent on beta-amyl- Schwarz AJ, Yu P, Miller BB, Shcherbinin S, Dickson J, Navitsky M,
oid in normal elderly. Ann Neurol 2013; 73: 472–80. et al. Regional profiles of the candidate tau PET ligand 18F-AV-
Lowe VJ, Curran G, Fang P, Liesinger AM, Josephs KA, Parisi JE, 1451 recapitulate key features of Braak histopathological stages.
et al. An autoradiographic evaluation of AV-1451 Tau PET in de- Brain 2016; 139 (Pt 5): 1539–50.
mentia. Acta Neuropathol Commun 2016; 4: 58. Sperling RA, Rentz DM, Johnson KA, Karlawish J, Donohue M,
Lowe VJ, Weigand SD, Senjem ML, Vemuri P, Jordan L, Kantarci K, Salmon DP, et al. The A4 study: stopping AD before symptoms
et al. Association of hypometabolism and amyloid levels in aging, begin? Sci Transl Med 2014; 6: 228fs13.
normal subjects. Neurology 2014; 82: 1959–67. Stancu IC, Vasconcelos B, Ris L, Wang P, Villers A, Peeraer E, et al.
Meltzer CC, Kinahan PE, Greer PJ, Nichols TE, Comtat C, Cantwell Templated misfolding of Tau by prion-like seeding along neuronal
MN, et al. Comparative evaluation of MR-based partial-volume connections impairs neuronal network function and associated be-
correction schemes for PET. J Nucl Med 1999; 40: 2053–65. havioral outcomes in Tau transgenic mice. Acta Neuropathol 2015;
Murray ME, Lowe VJ, Graff-Radford NR, Liesinger AM, Cannon A,
129: 875–94.
Przybelski SA, et al. Clinicopathologic and 11C-Pittsburgh com-
Takeda S, Wegmann S, Cho H, DeVos SL, Commins C, Roe AD, et al.
pound B implications of Thal amyloid phase across the
Neuronal uptake and propagation of a rare phosphorylated high-
Alzheimer’s disease spectrum. Brain 2015; 138 (Pt 5): 1370–81.
molecular-weight tau derived from Alzheimer’s disease brain. Nat
Nelson PT, Alafuzoff I, Bigio EH, Bouras C, Braak H, Cairns NJ,
Commun 2015; 6: 8490.
et al. Correlation of Alzheimer disease neuropathologic changes
Tzourio-Mazoyer N, Landeau B, Papathanassiou D, Crivello F, Etard
with cognitive status: a review of the literature. J Neuropathol
O, Delcroix N, et al. Automated anatomical labeling of activations
Exp Neurol 2012; 71: 362–81.
in SPM using a macroscopic anatomical parcellation of the MNI
Okamura N, Harada R, Furumoto S, Arai H, Yanai K, Kudo Y. Tau
MRI single-subject brain. Neuroimage 2002; 15: 273–89.
PET imaging in Alzheimer’s disease. Curr Neurol Neurosci Rep
Vemuri P, Gunter JL, Senjem ML, Whitwell JL, Kantarci K, Knopman
2014; 14: 500.
Ossenkoppele R, Pijnenburg YA, Perry DC, Cohn-Sheehy BI, Scheltens DS, et al. Alzheimer’s disease diagnosis in individual subjects using
NM, Vogel JW, et al. The behavioural/dysexecutive variant of structural MR images: validation studies. Neuroimage 2008a; 39:
Alzheimer’s disease: clinical, neuroimaging and pathological fea- 1186–97.
tures. Brain 2015; 138 (Pt 9): 2732–49. Vemuri P, Whitwell JL, Kantarci K, Josephs KA, Parisi JE, Shiung MS,
Ossenkoppele R, Schonhaut DR, Scholl M, Lockhart SN, Ayakta N, et al. Antemortem MRI based STructural Abnormality iNDex
Baker SL, et al. Tau PET patterns mirror clinical and neuroanatom- (STAND)-scores correlate with postmortem Braak neurofibrillary
ical variability in Alzheimer’s disease. Brain 2016; 139 (Pt 5): 1551– tangle stage. Neuroimage 2008b; 42: 559–67.
67. Xia CF, Arteaga J, Chen G, Gangadharmath U, Gomez LF, Kasi D,
Ossenkoppele R, Zwan MD, Tolboom N, van Assema DM, Adriaanse et al. [(18)F]T807, a novel tau positron emission tomography ima-
SF, Kloet RW, et al. Amyloid burden and metabolic function in ging agent for Alzheimer’s disease. Alzheimers Dement 2013a; 9:
early-onset Alzheimer’s disease: parietal lobe involvement. Brain 666–76.
2012; 135 (Pt 7): 2115–25. Xia M, Wang J, He Y. BrainNet viewer: a network visualization tool
Petersen RC. Mild cognitive impairment as a diagnostic entity. J Intern for human brain connectomics. PLoS One 2013b; 8: e68910.
Med 2004; 256: 183–94. Zheng WH, Bastianetto S, Mennicken F, Ma W, Kar S. Amyloid beta
Pontecorvo MJ, Devous MD Sr, Navitsky M, Lu M, Salloway S, peptide induces tau phosphorylation and loss of cholinergic neurons
Schaerf FW, et al. Relationships between flortaucipir PET tau bind- in rat primary septal cultures. Neuroscience 2002; 115: 201–11.
ing and amyloid burden, clinical diagnosis, age and cognition. Brain Zimmer ER, Leuzy A, Gauthier S, Rosa-Neto P. Developments in Tau
2017; 140: 748–63. PET Imaging. Can J Neurol Sci 2014; 41: 547–53.