Blood Phosphorylated Tau 181 as a Biomarker for Amyloid Burden on Brain PET in Cognitively Healthy Adults

Abstract

Background:

Plasma phosphorylated-tau181 (p-tau181) is a promising biomarker for Alzheimer’s disease (AD) and may offer utility for predicting preclinical disease.

Objective:

To evaluate the prospective association between plasma p-tau181 and amyloid-β (Aβ) and tau-PET deposition in cognitively unimpaired individuals.

Methods:

Plasma p-tau181 levels were measured at baseline in 52 [48% women, mean 64.4 (SD 5.5) years] cognitively unimpaired Framingham Offspring cohort participants using samples stored between 2011–2014 who subsequently underwent ¹¹C-Pittsburgh Compound-B (PiB)-PET and/or ¹⁸F-Flortaucipir (FTP)-PET scans (n = 18 with tau-PET) a mean of 6.8 (SD 0.6) years later. Our primary outcomes included Aβ-precuneus, Aβ-FLR (frontal, lateral, and retrosplenial cortices) and tau-global composite region PET deposition. Secondary outcomes included individual regional Aβ and tau PET-deposition. P-tau181 was compared with plasma neurofilament light chain (NFL) and glial fibrillary acidic protein (GFAP) in predicting PET outcomes.

Results:

P-tau181 was associated with increased Aβ deposition in the FLR (β±SE, 1.25±0.30, p < 0.0001), precuneus (1.35±0.29, p < 0.001), and other cortical regions. Plasma NFL (1.30±0.49, p = 0.01) and GFAP (1.46±0.39, p < 0.001) were also associated with FLR Aβ deposition. In models including all three biomarkers adjusted for age, sex, APOE E4 allele, AD polygenic risk score and cortical atrophy score, p-tau181 (0.93±0.31, p < 0.01, R² = 0.18) and GFAP (0.93±0.41, p = 0.03, R² = 0.11), but not NFL (0.25±0.51, p = 0.62, R² = 0.01), were associated with FLR-Aβ deposition. Plasma p-tau181 was not associated with tau-PET burden.

Conclusion:

In cognitively unimpaired adults, elevated plasma p-tau181 is associated with future increased Aβ deposition across multiple brain regions. Our results highlight the potential utility of p-tau181 as a blood-biomarker to screen for brain-amyloid deposition in cognitively healthy individuals in a community-setting.

Keywords

Dementia brain positron emission tomography Aβ-PET tau-PET biomarkers

INTRODUCTION

Compared to cerebrospinal fluid (CSF) or PET biomarkers of amyloid-β (Aβ) and tau, blood biomarkers are a more convenient option for dementia risk prediction, due to greater patient-acceptability, potential for wider scalability and lower cost. Plasma tau phosphorylated at threonine 181 (p-tau181) has been identified as a promising candidate biomarker for Alzheimer’s disease (AD), demonstrating greater specificity compared to circulating total-tau (t-tau) levels, and a strong correlation with CSF concentrations [1]. Plasma p-tau181 has been associated with global Aβ and tau burden on PET as well as cognitive decline and AD dementia in a number of recent studies [1 –9]. However, the role of p-tau181 as a potential biomarker of Aβ- and tau-burden on brain-PET in community-based, cognitively unimpaired individuals followed forward over time has been less well studied [10]. Many of the prior studies included individuals with an underlying genetic predisposition to AD, combined cognitively unimpaired individuals and those with mild cognitive impairment and clinical dementia, employed a cross-sectional design, or were subject to referral and selection biases inherent in case-control designs of clinic- and autopsy-selected cases. Furthermore, the role of p-tau181 in predicting future early regional Aβ and tau PET deposition in AD (e.g., precuneus-Aβ or entorhinal-tau [11]), and its comparative performance to other plasma biomarkers in predicting brain amyloid and tau deposition in community-based, cognitively healthy individuals has been incompletely explored (e.g., comparison with markers such as glial fibrillary acidic protein [GFAP]). Whether circulating levels of p-tau181 in the periphery are influenced by underlying vascular risk factors, also remains to be determined [3]. A greater understanding of the determinants of circulating p-tau181 concentrations and the role of p-tau181 in predicting brain amyloid and tau burden in cognitively healthy individuals, would greatly inform future approaches to preclinical risk prediction in asymptomatic, community-based adults.

In this investigation, we evaluated the prospective association between plasma p-tau181 concentrations and global and regional Aβ and tau deposition on brain-PET, and determinants of circulating phospho-tau181 concentrations, in a community-based sample of stroke-free, cognitively unimpaired adults.

MATERIALS AND METHODS

Study sample

The Framingham Heart Study (FHS) Offspring cohort is a longitudinal, community-based cohort of individuals enrolled between 1971 and 1975. Offspring cohort participants have been followed for the development of vascular risk factors, cardiovascular disease and adverse cognitive outcomes for more than 45 years via quadrennial examination visits [12]. Offspring cohort participants undergo routine monitoring for cognitive impairment at each examination visit. Participants with suspected cognitive impairment undergo an extensive neuropsychological test battery, and if flagged for possible cognitive impairment on this testing, are evaluated by a neurologist (further details have been published previously) [13]. For the purposes of this investigation, we included Offspring cohort participants who attended the 9^th examination cycle (2011–2014), had plasma p-Tau 181 measured at baseline from samples stored at this examination, underwent ¹¹C-Pittsburgh Compound-B (PiB)-PET (n = 52) and/or ¹⁸F-Flortaucipir (FTP)-PET scans (n = 18) between 2016 and 2019 and were confirmed to be free of stroke and dementia at the time of brain-PET. All participants provided written informed consent before taking part in this study and the study protocols and consent forms were approved by the Boston University Medical Center Institutional Review Board.

Outcome measures

Overall Aβ and tau deposition in the brain were measured using PiB-PET and FTP-PET, respectively. The PET acquisition and processing protocols have previously been described [14, 15] and have been validated against tissue Aβ burden at autopsy [16]. In brief, PET imaging was performed using a Siemens ECAT HR + scanner. PiB-PET images were obtained following a 10–15 mCi bolus injection over a 60-min dynamic acquisition period. FTP-PET images were obtained following a 9–11 mCi bolus injection over 80–100 min using a series of 4×5-min frames. Images were co-registered to T1-weighted structural MRI brain sequences using SPM8 and regions of interest (ROIs) were derived using FreeSurfer software v6.0 [17].

PiB-PET retention was reported as distribution volume ratios for each ROI based on the validated Logan graphical analysis technique [18, 19], and expressed relative to cerebellar cortex (due to its negligible specific binding of PiB) [20]. Our primary Aβ outcome measures included a global composite measure of PiB retention in the frontal, lateral, and retrosplenial regions (FLR), and the precuneus. FLR uptake was computed from mean PiB uptake in the superior frontal, inferior frontal, rostral middle frontal, rostral anterior cingulate, medial orbitofrontal, inferior temporal, middle temporal, inferior parietal, and precuneus regions [20]. We selected FLR uptake as our primary outcome as there is significant PiB uptake in this region in patients with diagnosed AD [21] and FLR has been used as a composite measure of global Aβ retention in multiple prior studies [22, 23]. We selected the precuneus as it is one of the earliest affected regions in AD and a marker of early, preclinical disease [24]. For secondary outcomes, we explored Aβ deposition within individual cortical regions (listed in Table 5).

FTP-PET retention was reported as standardised uptake value ratios (SUVr) using mean cerebellar grey matter retention as a reference. Our primary tau outcome measure consisted of a global composite measure of FTP uptake in the entorhinal cortex, fusiform gyrus and the inferior and middle temporal cortices, as these areas are susceptible to early tau deposition[25 –27]. Secondary outcomes included regional cortical tau deposition (Table 5). For each ROI, we averaged values for the left and right hemispheres. PET imaging data were not partial volume corrected due to the minimal degree of brain atrophy in our relatively young cohort.

Laboratory measurement of biomarkers

Blood biomarkers, namely p-tau181, neurofilament light chain (NFL), and GFAP were measured from blood samples obtained and stored at the baseline examination visit (9^th examination cycle). Early morning plasma samples were obtained in ethylenediaminetetraacetic acid vials from consenting participants who had been lying in the supine position for 10 min. Tubes were then immediately centrifuged, and the supernatant plasma was aliquoted within 30 min into 100μL aliquots and stored in a freezer at –80°C until the time of assay performance. Samples were transferred on dry ice to the University of Vermont where they were thawed for the first time at the time of assay run. Samples were analyzed using a highly sensitive Quanterix Single Molecule Array (Simoa) assay. Lower limits of detection were as follows: p-tau181, 0.112 pg/mL; NFL, 2.9 pg/mL; and GFAP, 4.6 pg/mL. For p-tau 181, NFL and GFAP, the inter-assay coefficients of variation (CV) were 6.2%, 10.2% and 10.2% and the intra-assay CV were 7.8%, 4.7%, and 2.7%, respectively. The lower limits of detection were consistent with manufacturer specified cutoffs.

Cohort characteristics

We measured baseline demographic and clinical covariates at the 9^th examination cycle, including age, sex, education (self-reported with the following categories: high school degree, some college experience but no college degree, college degree or higher), systolic blood pressure (SBP), diastolic blood pressure (DBP), use of antihypertensive medication, history of hypertension, prevalent cardiovascular disease (CVD, including peripheral vascular disease; coronary heart disease [coronary insufficiency, angina, myocardial infarction]; and cerebrovascular disease [transient ischemic attack and stroke]; and congestive heart failure), history of diabetes mellitus, body mass index, history of smoking, total cholesterol, high density lipoprotein cholesterol, use of lipid lowering medications, apolipoprotein E E4 (APOE4) carrier status, cortical AD signature score and a polygenic risk score for dementia based on several important risk loci for AD dementia [28]. The cortical AD signature score is a measure of the degree of cortical atrophy in typically affected regions in patients with AD and is proposed as a useful tool for prediction of AD (Satizabal CL, unpublished data). SBP and DBP were defined as the mean of two physician-recorded measurements obtained from the participant’s left arm performed in the seated position. Current smoking was defined as participant self-reported smoking within the previous 12 months. Diabetes mellitus was defined as a fasting blood glucose≥7 mmol/L, a random blood glucose≥11.1 mmol/L, or use of insulin or oral hypoglycemic medications.

Statistical analysis

Levels of p-tau181, GFAP, and NFL and FLR-Aβ PET deposition were logarithmically transformed to normalize their distributions and all PET outcome data were standardized prior to analyses. For our primary analyses, we used linear regression models to evaluate the association between plasma p-tau181 and Aβ (FLR and precuneus) and tau (global composite measure) PET deposition, first adjusting for age, sex, and time from blood sample collection to completion of PET-brain (model 1), and if significant, additionally adjusting for APOE E4 carrier status, polygenic risk score for AD, a cortical AD signature score and plasma NFL and GFAP (model 2). We calculated the proportion of variability explained (R²) in the outcome on addition of each biomarker to the full multivariable model.

We generated a stepwise model to identify the strongest predictors of Aβ and/or tau PET retention using the above listed dependent variables, with age and sex automatically retained in the model. We used an unadjusted linear regression model to identify covariates associated with circulating concentrations of p-tau181 in the periphery. In secondary analyses, we evaluated the association between p-tau181 and regional cortical Aβ and tau PET retention. In an exploratory analysis, we tested for an interaction between p-tau181 and sex in their association FLR PET.

For all statistical analyses, p < 0.05 was considered significant. Analyses were completed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA).

RESULTS

Our sample consisted of 52 participants with available Aβ PET and biomarker data. The mean age was 64 years (SD 6) and 48% of participants were women. The mean time between blood sample collection and brain-PET was 6.8 (SD 0.6) years. The median baseline p-tau181 level was 1.5 pg/mL (Q1–Q3, 1.2–1.9). Baseline characteristics are shown in Table 1.

Table 1

Baseline characteristics

Variable	Overall
	(n = 52)
No. (%)
Age, y, mean (SD)	64.4 (5.5)
Women	25 (48.1)
Education
High school degree	8 (15.4)
Some years of college	10 (19.2)
College degree or higher	34 (65.4)
Systolic blood pressure, mmHg, mean (SD)	118.3 (13.0)
Diastolic blood pressure, mmHg, mean (SD)	72.1 (8.4)
Hypertension	26 (50.0)
Prevalent CVD	8 (15.4)
BMI, kg/m², mean (SD)	28.6 (5.2)
Anti-hypertensive medication	24 (46.2)
Lipid-lowering medication	28 (53.9)
Current smoker	2 (3.9)
ApoE4 allele carrier	10 (19.2)
Diabetes mellitus	7 (13.5)
TC, mmol/L, mean (SD)	183.1 (40.8)
HDL-C, mmol/L, mean (SD)	60.3 (17.8)
p-Tau181, pg/mL, median (Q1, Q3)	1.5 (1.2, 1.9)
NFL, pg/mL, median (Q1, Q3)	13.5 (10.4, 17.2)
GFAP, pg/mL, median (Q1, Q3)	138.8 (108.3-178.2)
Log p-Tau 181, mean (SD)	0.4 (0.4)
Log NFL, mean (SD)	2.6 (0.3)
Log GFAP, mean (SD)	5.0 (0.4)

Values are reported as n (%), unless otherwise specified. Baseline demographic and clinical characteristics were defined at examination 9. Biomarker values were logarithmically transformed to approximate a normal distribution. SD, standard deviation; CVD, cardiovascular disease; BMI, body mass index; APOE E4, apolipoprotein E4 allele; TC, total cholesterol; HDL-C, high density lipoprotein cholesterol; p-Tau181, phosphorylated-tau 181; NFL, neurofilament light chain; GFAP, glial fibrillary acidic protein.

Plasma p-tau181 and Aβ PET deposition

In multivariable models adjusted for age, sex, and time from blood sample collection to PET, elevated p-tau181 was associated with increased Aβ deposition in the FLR (β±SE, 1.25±0.30, p < 0.001) and precuneus regions (1.35±0.30, p < 0.001). In multivariable models additionally adjusted for APOE E4 allele status, AD polygenic risk score, a cortical atrophy score and plasma NFL and GFAP concentrations, elevated p-tau181 remained associated with Aβ deposition in the FLR (0.93±0.31, p < 0.01) and precuneus (1.01±0.30, p < 0.01) regions (Table 2). In multivariable models adjusted for age, sex, and time from biomarker measurement to PET, circulating p-tau181 was associated with Aβ deposition in multiple cortical regions, including the medial orbitofrontal (1.07±0.31, p = 0.001), rostral middle frontal (1.03±0.31, p = 0.002), superior fontal (1.10±0.31, p = 0.001), superior temporal (1.24±0.30, p < 0.001), middle temporal (1.19±0.30, p < 0.001), and inferior temporal regions (1.01±0.32, p = 0.003) (Table 3). There was no significant interaction between p-tau181 and sex in their association with FLR-PET (p = 0.272).

Table 2

P-tau181 and Aβ PET burden

	Aβ
	FLR (SDU)		Precuneus (SDU)
	B±SE	p	B±SE	p
Model 1
p-tau181	1.25±0.30	< 0.001	1.35±0.29	< 0.001
NFL	1.30±0.49	0.01	1.44±0.49	< 0.01
GFAP	1.46±0.39	< 0.001	1.49±0.40	< 0.001
Model 2
p-tau181	0.93±0.31	< 0.01	1.01±0.30	< 0.01
NFL	0.25±0.51	0.62	0.29±0.49	0.56
GFAP	0.93±0.41	0.03	0.91±0.39	0.03
Age	–0.01±0.03	0.74	–0.02±0.03	0.44
Male	–0.30±0.27	0.26	–0.21±0.26	0.42
Time to PET	0.56±0.22	0.01	0.53±0.21	0.02
APOE E4 carrier	0.14±0.38	0.72	0.10±0.37	0.79
PGRS	–1.69±19.37	0.93	2.90±18.81	0.88
Cortical AD signature	–1.03±0.62	0.10	–1.22±0.60	0.05

Values for p-tau181 and FLR PET were logarithmically transformed to approximate a normal distribution. Model 1: Adjusted for age, sex, and time from blood sample collection to PET. Model 2: Adjusted for age, sex, time from blood sample collection to PET, APOE E4 carrier status, polygenic risk score for AD, cortical AD signature, and the remaining two biomarkers (e.g., the estimate for NFL is adjusted for GFAP and p-tau181). SDU, standard deviation units; SE, standard error; FLR, frontal, lateral and retrosplenial cortices; PET, positron emission tomography; PGRS, polygenic risk score; AD, Alzheimer’s disease; NFL, neurofilament light chain; GFAP, glial fibrillary acidic protein. n = 52 with Aβ-PET.

Table 3

Circulating p-tau181 concentrations and regional Aβ and global and regional tau deposition on brain-PET

	Log p-tau181
Aβ	B±SE	p
Medial orbitofrontal	1.07±0.31	0.001
Lateral orbitofrontal	1.11±0.30	< 0.001
Caudal middle frontal	1.14±0.31	< 0.001
Rostral middle frontal	1.03±0.31	0.002
Superior frontal	1.10±0.31	0.001
Frontal pole	0.85±0.32	0.01
Inferior temporal	1.01±0.32	0.003
Middle temporal	1.19±0.30	< 0.001
Superior temporal	1.24±0.30	< 0.001
Fusiform	1.10±0.32	0.001
Temporal pole	0.70±0.35	0.05
Insula	1.17±0.31	< 0.001
Isthmus	1.08±0.32	0.002
Tau
Tau composite	0.05±0.64	0.94
Entorhinal	–0.31±0.63	0.63
Inferior temporal	–0.19±0.64	0.77
Precuneus	–0.21±0.66	0.75

Values for p-tau181 were logarithmically transformed to approximate a normal distribution. Models are adjusted for age, sex and time from blood sample collection to performance of PET scan. Tau composite included mean FTP uptake in the entorhinal cortex, fusiform gyrus and the inferior and middle temporal cortices. SE, standard error. n = 52 with Aβ-PET, n = 18 with tau-PET.

Plasma p-tau181 and tau PET deposition

Plasma p-tau181 was not associated with overall tau PET deposition (global composite region: 0.05±0.64, p = 0.94) or with tau-PET deposition in any of the examined regions, including the entorhinal cortex, inferior temporal cortex, and the precuneus (n = 18 with available FTP-PET) (Table 3).

Plasma p-tau181 compared to plasma NFL and GFAP

In models adjusted for age, sex, and time from blood sample collection to PET, plasma NFL, and GFAP were each independently associated with Aβ deposition in the FLR (1.30±0.49, p = 0.01 and 1.46±0.39, p < 0.001, respectively) and precuneus regions (1.44±0.49, p < 0.01 and 1.49±0.40, p < 0.001, respectively). In fully adjusted models, p-tau181 (0.93±0.31, p < 0.01, R² = 0.18) and GFAP (0.93±0.41, p = 0.03, R² = 0.11), but not NFL (0.25±0.51, p = 0.62, R² = 0.01), remained associated with FLR Aβ deposition, with similar findings in the precuneus (Table 2). Almost 20% of the variability in FLR Aβ deposition was explained by plasma p-tau181 levels (R² = 0.18), while plasma GFAP levels explained 11% of the variability (Table 2). On stepwise analysis, p-tau181 (0.99±0.29, p = 0.001, R² = 0.20) and GFAP (1.07±0.37, p < 0.01, R² = 0.16) were identified as the most important predictors of Aβ FLR deposition, explaining 20% and 16% of the variability in Aβ FLR deposition, respectively (Table 4). In the precuneus region, a more sensitive measure of early Aβ deposition, P-tau181 (R² = 0.26)m and GFAP (R² = 0.14) together explained 40% of the variability in Aβ deposition.

Table 4

Stepwise model: Predictors of Aβ PET burden

	Aβ
	FLR (SDU)		Precuneus (SDU)
	B±SE	p	B±SE	p
p-tau181	0.99±0.29	0.001	1.09±0.27	< 0.001
NFL	–	–	–	–
GFAP	1.07±0.37	0.006	0.95±0.35	0.01
APOE E4 carrier	–	–	–	–
PGRS	–	–	–	–
Cortical AD signature	–	–	–1.32±0.60	0.03

Values for p-tau181, NFL, GFAP, and FLR PET were logarithmically transformed to approximate a normal distribution. All estimates are adjusted for age, sex and time from blood sample collection to performance of PET scan. SDU, standard deviation units; SE, standard error; FLR, frontal, lateral and retrosplenial cortices; PET, positron emission tomography; PGRS, polygenic risk score; AD, Alzheimer’s disease; NFL, neurofilament light chain; GFAP, glial fibrillary acidic protein.

Determinants of circulating p-tau 181

In age and sex-adjusted models, prevalent CVD, diabetes mellitus and other vascular risk factors were not associated with plasma p-tau181 (Table 5).

Table 5

Determinants of circulating p-tau181 concentrations

	Log p-tau181
	B±SE	p
Systolic blood pressure	–0.001±0.005	0.88
Diastolic blood pressure	–0.005±0.008	0.50
Antihypertensive therapy	0.16±0.11	0.17
BMI	0.02±0.01	0.18
Current smoking	–0.38±0.31	0.23
TC/HDL-C	0.15±0.12	0.21
Lipid-lowering medication	–0.08±0.07	0.27
Education
High school degree	–0.07±0.17	0.67
Some years of college	0.03±0.16	0.84
College degree or higher	ref	ref
Prevalent CVD	0.16±0.18	0.36
Prevalent diabetes mellitus	–0.04±0.17	0.81

Values for p-tau181 were logarithmically transformed to approximate a normal distribution. All estimates are adjusted for age and sex. SE, standard error; CVD, cardiovascular disease; BMI, body mass index; TC, total cholesterol; HDL-C, high density lipoprotein cholesterol; APOE E4, apolipoprotein E4.

DISCUSSION

Elevated plasma p-tau181 was prospectively associated with early Aβ deposition across multiple cortical regions in this sample of community-based, cognitively unimpaired adults. Circulating p-tau181 concentrations were independent of vascular risk factors.

In our investigation, circulating p-tau181 was associated with global Aβ deposition, as well as Aβ deposition in the precuneus, one of the earliest affected regions in AD. Our findings extend those from recent studies [1 –6], by demonstrating a prospective association between elevated plasma p-tau181 concentrations in cognitively healthy asymptomatic adults and Aβ PET deposition on average seven years later. Elevated levels of plasma p-tau181 have previously been reported in patients with autosomal dominant AD up to 16 years before the onset of clinical dementia [7], further supporting an important role for plasma p-tau181 as an early marker of presymptomatic disease.

The association between plasma p-tau181 and Aβ deposition in our sample involved multiple cortical areas throughout the frontal, parietal, and temporal lobes. Two recent studies also reported a correlation between elevated plasma p-tau181 levels and greater Aβ-PET deposition across several cortical regions including the temporal, parietal and frontal cortices [6], particularly, the precuneus and frontal regions [1], consistent with our findings. We hypothesized that APOE E4 carrier status may mediate the association between p-tau181 and Aβ-PET deposition. However, adjusting for APOE E4 carrier status did not materially alter our results. Furthermore, in a previous study p-tau181 demonstrated comparable accuracy in predicting Aβ-PET burden, regardless of underlying APOE E4 allele carrier status, similar to our findings [1]. Overall, plasma p-tau181 appears to be a useful marker of future global and regional Aβ burden on PET which appears to be unrelated to APOE E4 status.

Interestingly, we observed no association between circulating p-tau181 and either global or regional tau deposition on brain-PET in our sample of cognitively unimpaired individuals. Under the amyloid cascade hypothesis, neocortical tau deposition occurs downstream to the development of Aβ pathology, consistent with our finding of a lack of an association between p-tau181 and tau-PET in asymptomatic individuals [29]. P-tau181 may be a marker of preclinical AD pathology, reflecting early intraneuronal changes in response to initial deposits of Aβ in the brain, prior to the accumulation of significant tau pathology [30, 31]. In a recent study, p-tau181 was significantly correlated with Aβ-PET (r = 0.41, p = 0.0001) but not tau-PET (r = 0.14, p = 0.09) in the subgroup of cognitively unimpaired individuals, consistent with our findings, with high plasma p-tau181 levels noted in Aβ⁺ but tau^- PET cases [1]. Similarly in another study, plasma p-tau181 levels were associated with Aβ-PET but not tau-PET in cognitively unimpaired individuals (n = 172) [3]. However, a third study reported conflicting results, with a significant correlation between plasma p-tau181 levels and future tau-PET signal in the medial temporal and posterior cingulate cortices in cognitively healthy adults [6]. Importantly, our sample included only 18 individuals with available FTP-PET data, thus our tau-PET results, while consistent with a number of recent studies, should be interpreted with caution due to the small sample size.

In our investigation, plasma NFL and GFAP were each independently associated with Aβ burden on brain-PET. In multivariable models incorporating NFL and GFAP in addition to p-tau181 and other traditional risk factors (age, sex, APOE E4, etc.), p-tau181 was the strongest biomarker and best overall predictor of global Aβ burden on brain-PET, predicting 18% of the variability in Aβ-FLR burden. Plasma GFAP explained 11% of the variability in Aβ FLR but plasma NFL was no longer significant after multivariable adjustment. In a recent study, p-tau181 was found to be a stronger predictor of AD than traditional risk factors (including age and APOE E4 status) or other plasma biomarkers (t-tau, Aβ₄₂, and Aβ_42/20) [1], while in another study, plasma p-tau181 outperformed plasma Aβ₄₂/Aβ₄₀ ratio and NFL in predicting AD dementia risk amongst cognitively unimpaired individuals, consistent with our findings [4].

In an age- and sex-adjusted stepwise model to identify the most important predictors of early brain Aβ deposition, p-tau181, and GFAP were the two most important predictors of Aβ FLR burden, together explaining over one-third of the variability in brain Aβ burden. In the precuneus region, a more sensitive measure of early Aβ deposition, p-tau181, and GFAP explained 40% of the variability in Aβ deposition. Plasma NFL, a marker of axonal degeneration, was not a significant predictor of Aβ-PET burden in our sample of cognitively healthy adults. Compared to plasma NFL, p-tau181 and GFAP may be more useful biomarkers of early Aβ deposition, reflecting early intraneuronal changes and astrocytic responses, while NFL may offer greater utility as a biomarker further along the AD clinical continuum, e.g., prognostically in predicting rate of disease progression [32]. Compared to both NFL and GFAP, p-tau181 offers the advantage of greater specificity, showing selectively elevated levels in patients with AD but not other neurodegenerative diseases [4 , 33]. Plasma GFAP and NFL have been reported to be elevated in a number of other neurodegenerative conditions, including subtypes of frontotemporal dementia, primary progressive aphasia, corticobasal syndrome, and atypical Parkinsonian syndromes [34 –36].

In our sample, vascular risk factors or a history of vascular events were not associated with circulating p-tau181 levels. To our knowledge, there have been no prior studies exploring determinants of circulating p-tau181 levels in the periphery. Circulating p-tau181 appears to be a relatively robust biomarker for early Aβ deposition in a community-based sample of cognitively healthy adults, regardless of underlying vascular risk profiles. Previous studies have reported low intra-individual variability in plasma p-tau181 levels [2] which further supports its clinical utility. Plasma p-tau181 offers potential utility as a biomarker for preclinical dementia in several clinical contexts, including as part of a combined clinical/biomarker risk prediction tool for AD in primary care, as a simple diagnostic test for early disease detection, and as a screening tool and cost-saving strategy for enrolling suitable participants in clinical trials (reducing proportion of negative PET scans by excluding those with subthreshold plasma p-tau181 concentrations).

Strengths of our study include the use of a sample of community-based individuals confirmed to be stroke-free and cognitively unimpaired at baseline, in-depth clinical and plasma biomarker phenotyping of our cohort, use of Aβ- and tau-PET (gold standard surrogate markers for preclinical dementia), and the relatively long duration of follow-up compared to other studies to date (mean 7 years). Our study has some important limitations. Our sample is predominantly Caucasian, which may limit the generalizability of our findings to other ethnicities. Further studies in more diverse cohorts are required to validate our findings. The subsample of individuals with available PET neuroimaging data was relatively small, and only 18 had available FTP-PET, thus our results on tau-PET should be interpreted with caution. We are also unable to comment on the comparative performance of plasma p-tau181 in Aβ-PET positive versus negative individuals, as none of the included participants in our sample were amyloid-PET positive based on clinical criteria (visual reads). This is likely due to the younger age and cognitively healthy status of our sample at baseline. Furthermore, we did not have access to PET neuroimaging data at baseline to study the association between p-tau181 and longitudinal changes in Aβ and tau PET burden. We also did not have access to data on CSF levels of p-tau181 in our cohort, however our focus was on an accessible, easy-to-measure blood biomarker not necessitating invasive lumbar puncture testing.

Conclusions

Elevated plasma p-tau181 is prospectively associated with early cortical Aβ deposition in a community-based sample of cognitively unimpaired adults, validating and extending the findings from recent studies. Circulating p-tau181 concentrations appear to be independent of most vascular risk factors, highlighting its potential robustness and generalizability as a biomarker in a primary-care setting. Compared to traditional risk factors and other plasma biomarkers, p-tau181 was the strongest biomarker and best overall predictor of Aβ burden on brain-PET. Circulating p-tau181 offers potential utility as a risk stratifying tool for enrolling participants for clinical trials of early disease modifiers and for population-level screening for early Aβ deposition on brain-PET.

Footnotes

ACKNOWLEDGMENTS

The authors would like to acknowledge the contribution of the Framingham Heart Study participants.

This research was supported by the Health Research Board of Ireland (CSF-2020-011) and the Alzheimer’s Association (AACSF-18-566570). The Framingham Heart Study is supported by the NHLBI (contract no. N01-HC-25195, no. HHSN268201500001, and no. 75N92019D00031). This research was also supported by NHLBI grants R01 HL60040 and R01 HL70100, grants from the National Institute on Aging (R01 AG054076, R01 AG049607, R01 AG033193, U01 AG049505, U01 AG052409) and grants from the National Institute of Neurological Disorders and Stroke (NS017950 and UH2 NS100605).

None of the funding entities had any role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.

Dr. McGrath had full access to all the data in the study and takes responsibility for its integrity and the data analysis.

Authors’ disclosures available online ().

References

Karikari

, Pascoal

, Ashton

, Janelidze

, Benedet

, Rodriguez

, Chamoun

, Savard

, Kang

, Therriault

, Schöll

, Massarweh

, Soucy

, Höglund

, Brinkmalm

, Mattsson

, Palmqvist

, Gauthier

, Stomrud

, Zetterberg

, Hansson

, Rosa-Neto

, Blennow

(2020) Blood phosphorylated tau 181 as a biomarker for Alzheimer’s disease: A diagnostic performance and prediction modelling study using data from four prospective cohorts. Lancet Neurol 19, 422–433.

Karikari

, Benedet

, Ashton

, Lantero Rodriguez

, Snellman

, Suárez-Calvet

, Saha-Chaudhuri

, Lussier

, Kvartsberg

, Rial

, Pascoal

, Andreasson

, Schöll

, Weiner

, Rosa-Neto

, Trojanowski

, Shaw

, Blennow

, Zetterberg

(2021) Diagnostic performance and prediction of clinical progression of plasma phospho-tau181 in the Alzheimer’s Disease Neuroimaging Initiative. Mol Psychiatry 26, 429–442.

Mielke

, Hagen

, Xu

, Chai

, Vemuri

, Lowe

, Airey

, Knopman

, Roberts

, Machulda

, Jack

Jr. , Petersen

, Dage

(2018) Plasma phospho-tau181 increases with Alzheimer’s disease clinical severity and is associated with tau- and amyloid-positron emission tomography. Alzheimers Dement 14, 989–997.

Janelidze

, Mattsson

, Palmqvist

, Smith

, Beach

, Serrano

, Chai

, Proctor

, Eichenlaub

, Zetterberg

, Blennow

, Reiman

, Stomrud

, Dage

, Hansson

(2020) Plasma P-tau181 in Alzheimer’s disease: Relationship to other biomarkers, differential diagnosis, neuropathology and longitudinal progression to Alzheimer’s dementia. Nat Med 26, 379–386.

Thijssen

, La Joie

, Wolf

, Strom

, Wang

, Iaccarino

, Bourakova

, Cobigo

, Heuer

, Spina

, VandeVrede

, Chai

, Proctor

, Airey

, Shcherbinin

, Duggan Evans

, Sims

, Zetterberg

, Blennow

, Karydas

, Teunissen

, Kramer

, Grinberg

, Seeley

, Rosen

, Boeve

, Miller

, Rabinovici

, Dage

, Rojas

, Boxer

(2020) Diagnostic value of plasma phosphorylated tau181 in Alzheimer’s disease and frontotemporal lobar degeneration. Nat Med 26, 387–397.

Moscoso

, Grothe

, Ashton

, Karikari

, Rodriguez

, Snellman

, Suárez-Calvet

, Zetterberg

, Blennow

, Schöll

(2021) Time course of phosphorylated-tau181 in blood across the Alzheimer’s disease spectrum. Brain 144, 325–339.

O’Connor

, Karikari

, Poole

, Ashton

, Lantero Rodriguez

, Khatun

, Swift

, Heslegrave

, Abel

, Chung

, Weston

PSJ

, Pavisic

, Ryan

, Barker

, Rossor

, Polke

, Frost

, Mead

, Blennow

, Zetterberg

, Fox

(2021) Plasma phospho-tau181 in presymptomatic and symptomatic familial Alzheimer’s disease: A longitudinal cohort study. Mol Psychiatry 26, 5967–5976.

Cullen

, Leuzy

, Palmqvist

, Janelidze

, Stomrud

, Pesini

, Sarasa

, Allué

, Proctor

, Zetterberg

, Dage

, Blennow

, Mattsson-Carlgren

, Hansson

(2021) Individualized prognosis of cognitive decline and dementia in mild cognitive impairment based on plasma biomarker combinations. Nat Aging 1, 114–123.

Therriault

, Benedet

, Pascoal

, Lussier

, Tissot

, Karikari

, Ashton

, Chamoun

, Bezgin

, Mathotaarachchi

, Gauthier

, Saha-Chaudhuri

, Zetterberg

, Blennow

, Rosa-Neto

(2021) Association of plasma P-tau181 with memory decline in non-demented adults. Brain Commun 3, fcab136.

10.

Rauchmann

, Schneider-Axmann

, Perneczky

(2021) Associations of longitudinal plasma p-tau181 and NfL with tau-PET, Aβ-PET and cognition. J Neurol Neurosurg Psychiatry 92, 1289–1295.

11.

Schöll

, Maass

, Mattsson

, Ashton

, Blennow

, Zetterberg

, Jagust

(2019) Biomarkers for tau pathology. Mol Cell Neurosci 97, 18–33.

12.

Feinleib

, Kannel

, Garrison

, McNamara

, Castelli

(1975) The Framingham Offspring Study. Design and preliminary data. Prev Med 4, 518–525.

13.

Satizabal

, Beiser

, Seshadri

(2016) Incidence of dementia over three decades in the Framingham Heart Study. N Engl J Med 375, 93–94.

14.

Becker

, Hedden

, Carmasin

, Maye

, Rentz

, Putcha

, Fischl

, Greve

, Marshall

, Salloway

, Marks

, Buckner

, Sperling

, Johnson

(2011) Amyloid-β associated cortical thinning in clinically normal elderly. Ann Neurol 69, 1032–1042.

15.

Gatchel

, Donovan

, Locascio

, Schultz

, Becker

, Chhatwal

, Papp

, Amariglio

, Rentz

, Blacker

, Sperling

, Johnson

, Marshall

(2017) Depressive symptoms and tau accumulation in the inferior temporal lobe and entorhinal cortex in cognitively normal older adults: A pilot study. J Alzheimers Dis 59, 975–985.

16.

Ikonomovic

, Klunk

, Abrahamson

, Mathis

, Price

, Tsopelas

, Lopresti

, Ziolko

, Bi

, Paljug

, Debnath

, Hope

, Isanski

, Hamilton

, DeKosky

(2008) Post-mortem correlates of in vivo PiB-PET amyloid imaging in a typical case of Alzheimer’s disease. Brain 131, 1630–1645.

17.

Desikan

, Ségonne

, Fischl

, Quinn

, Dickerson

, Blacker

, Buckner

, Dale

, Maguire

, Hyman

, Albert

, Killiany

(2006) An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. Neuroimage 31, 968–980.

18.

Logan

, Alexoff

, Fowler

(2011) The use of alternative forms of graphical analysis to balance bias and precision in PET images. J Cereb Blood Flow Metab 31, 535–546.

19.

Price

, Klunk

, Lopresti

, Lu

, Hoge

, Ziolko

, Holt

, Meltzer

, DeKosky

, Mathis

(2005) Kinetic modeling of amyloid binding in humans using PET imaging and Pittsburgh Compound-B. J Cereb Blood Flow Metab 25, 1528–1547.

20.

Hanseeuw

, Betensky

, Mormino

, Schultz

, Sepulcre

, Becker

, Jacobs

HIL

, Buckley

, LaPoint

, Vannini

, Donovan

, Chhatwal

, Marshall

, Papp

, Amariglio

, Rentz

, Sperling

, Johnson

(2018) PET staging of amyloidosis using striatum. Alzheimers Dement 14, 1281–1292.

21.

Raji

, Becker

, Tsopelas

, Price

, Mathis

, Saxton

, Lopresti

, Hoge

, Ziolko

, DeKosky

, Klunk

(2008) Characterizing regional correlation, laterality and symmetry of amyloid deposition in mild cognitive impairment and Alzheimer’s disease with Pittsburgh Compound B. J Neurosci Methods 172, 277–282.

22.

Johnson

, Gregas

, Becker

, Kinnecom

, Salat

, Moran

, Smith

, Rosand

, Rentz

, Klunk

, Mathis

, Price

, Dekosky

, Fischman

, Greenberg

(2007) Imaging of amyloid burden and distribution in cerebral amyloid angiopathy. Ann Neurol 62, 229–234.

23.

Gonzales

, Samra

, O’Donnell

, Mackin

, Salinas

, Jacob

, Satizabal

, Aparicio

, Thibault

, Sanchez

, Finney

, Rubinstein

, Mayblyum

, Killiany

, Decarli

, Johnson

, Beiser

, Seshadri

(2021) Association of midlife depressive symptoms with regional amyloid-β and tau in the Framingham Heart Study. J Alzheimers Dis 82, 249–260.

24.

Vlassenko

, Benzinger

, Morris

(2012) PET amyloid-beta imaging in preclinical Alzheimer’s disease. Biochim Biophys Acta 1822, 370–379.

25.

Johnson

, Schultz

, Betensky

, Becker

, Sepulcre

, Rentz

, Mormino

, Chhatwal

, Amariglio

, Papp

, Marshall

, Albers

, Mauro

, Pepin

, Alverio

, Judge

, Philiossaint

, Shoup

, Yokell

, Dickerson

, Gomez-Isla

, Hyman

, Vasdev

, Sperling

(2016) Tau positron emission tomographic imaging in aging and early Alzheimer disease. Ann Neurol 79, 110–119.

26.

Hanseeuw

, Betensky

, Jacobs

HIL

, Schultz

, Sepulcre

, Becker

, Cosio

DMO

, Farrell

, Quiroz

, Mormino

, Buckley

, Papp

, Amariglio

, Dewachter

, Ivanoiu

, Huijbers

, Hedden

, Marshall

, Chhatwal

, Rentz

, Sperling

, Johnson

(2019) Association of amyloid and tau with cognition in preclinical Alzheimer disease: A longitudinal study. JAMA Neurol 76, 915–924.

27.

Sanchez

, Becker

, Jacobs

HIL

, Hanseeuw

, Jiang

, Schultz

, Properzi

, Katz

, Beiser

, Satizabal

, O’Donnell

, DeCarli

, Killiany

, Fakhri

, Normandin

, Gómez-Isla

, Quiroz

, Rentz

, Sperling

, Seshadri

, Augustinack

, Price

, Johnson

(2021) The cortical origin and initial spread of medial temporal tauopathy in Alzheimer’s disease assessed with positron emission tomography. Sci Transl Med 13, eabc0655.

28.

Kunkle

, Grenier-Boley

, Sims

, Bis

, Damotte

, Naj

, Boland

, Vronskaya

, van der Lee

, Amlie-Wolf

, et al. (2019) Genetic meta-analysis of diagnosed Alzheimer’s disease identifies new risk loci and implicates Aβ, tau, immunity and lipid processing. Nat Genet 51, 414–430.

29.

Hardy

, Selkoe

(2002) The amyloid hypothesis of Alzheimer’s disease: Progress and problems on the road to therapeutics. Science 297, 353–356.

30.

Sato

, Barthélemy

, Mawuenyega

, Patterson

, Gordon

, Jockel-Balsarotti

, Sullivan

, Crisp

, Kasten

, Kirmess

, Kanaan

, Yarasheski

, Baker-Nigh

, Benzinger

TLS

, Miller

, Karch

, Bateman

(2018) Tau kinetics in neurons and the human central nervous system. Neuron 97, 1284–1298.e1287.

31.

Mattsson-Carlgren

, Andersson

, Janelidze

, Ossenkoppele

, Insel

, Strandberg

, Zetterberg

, Rosen

, Rabinovici

, Chai

, Blennow

, Dage

, Stomrud

, Smith

, Palmqvist

, Hansson

(2020) Aβ deposition is associated with increases in soluble and phosphorylated tau that precede a positive Tau PET in Alzheimer’s disease. Sci Adv 6, eaaz2387.

32.

Preische

, Schultz

, Apel

, Kuhle

, Kaeser

, Barro

, Gräber

, Kuder-Buletta

, LaFougere

, Laske

, Vöglein

, Levin

, Masters

, Martins

, Schofield

, Rossor

, Graff-Radford

, Salloway

, Ghetti

, Ringman

, Noble

, Chhatwal

, Goate

, Benzinger

TLS

, Morris

, Bateman

, Wang

, Fagan

, McDade

, Gordon

, Jucker

(2019) Serum neurofilament dynamics predicts neurodegeneration and clinical progression in presymptomatic Alzheimer’s disease. Nat Med 25, 277–283.

33.

Grothe

, Moscoso

, Ashton

, Karikari

, Lantero-Rodriguez

, Snellman

, Zetterberg

, Blennow

, Schöll

, Alzheimer’s Disease Neuroimaging Initiative (2021) Associations of fully automated CSF and novel plasma biomarkers with Alzheimer disease neuropathology at autopsy. Neurology 97, e1229–e1242.

34.

Heller

, Foiani

, Moore

, Convery

, Bocchetta

, Neason

, Cash

, Thomas

, Greaves

, Woollacott

, Shafei

, Van Swieten

, Moreno

, Sanchez-Valle

, Borroni

, Laforce

Jr. , Masellis

, Tartaglia

, Graff

, Galimberti

, Rowe

, Finger

, Synofzik

, Vandenberghe

, de Mendonca

, Tagliavini

, Santana

, Ducharme

, Butler

, Gerhard

, Levin

, Danek

, Frisoni

, Sorbi

, Otto

, Heslegrave

, Zetterberg

, Rohrer

(2020) Plasma glial fibrillary acidic protein is raised in progranulin-associated frontotemporal dementia. J Neurol Neurosurg Psychiatry 91, 263–270.

35.

Heller

, Chan

, Foiani

, Todd

, Russell

, Greaves

, Heslegrave

, Warren

, Zetterberg

, Bocchetta

, Rohrer

(2020) Plasma glial fibrillary acidic protein and neurofilament light chain are measures of disease severity in semantic variant primary progressive aphasia. J Neurol Neurosurg Psychiatry, doi: 10.1136/jnnp-2020-325085.

36.

Hansson

, Janelidze

, Hall

, Magdalinou

, Lees

, Andreasson

, Norgren

, Linder

, Forsgren

, Constantinescu

, Zetterberg

, Blennow

(2017) Blood-based NfL: A biomarker for differential diagnosis of parkinsonian disorder. Neurology 88, 930–937.