Plasma Phosphorylated Tau 231 Increases at One-Year Intervals in Cognitively Unimpaired Subjects

Abstract

Background:

Plasma biomarkers of Alzheimer’s disease (AD) constitute a non-invasive tool for diagnosing and classifying subjects. They change even in preclinical stages, but it is necessary to understand their properties so they can be helpful in a clinical context.

Objective:

With this work we want to study the evolution of p-tau231 plasma levels in the preclinical stages of AD and its relationship with both cognitive and imaging parameters.

Methods:

We evaluated plasma phosphorylated (p)-tau231 levels in 146 cognitively unimpaired subjects in sequential visits. We performed a Linear Mixed-effects Model to analyze their rate of change. We also correlated their baseline levels with cognitive tests and structural and functional image values. ATN status was defined based on cerebrospinal fluid biomarkers.

Results:

Plasma p-tau231 showed a significant rate of change over time. It correlated negatively with memory tests only in amyloid-positive subjects. No significant correlations were found with any imaging measures.

Conclusions:

Increases in plasma p-tau231 can be detected at one-year intervals in cognitively healthy subjects. It could constitute a sensitive marker for detecting early signs of neuronal network impairment by amyloid.

Keywords

Alzheimer’s disease longitudinal study p-tau231 plasma biomarkers presymptomatic stages

INTRODUCTION

Alzheimer’s disease (AD) is a neurodegenerative process characterized by the accumulation of extracellular plaques of amyloid-β (Aβ) and intracellular neurofibrillary tangles of phosphorylated tau (p-tau) in the brain. These pathological changes happen decades before the first symptoms and are reflected in the cerebrospinal fluid (CSF) [1]. Thus, along the AD continuum, the concentration of CSF total tau (t-tau) and p-tau levels increase, and both Aβ₄₂ and Aβ₄₂/Aβ₄₀ ratio levels decrease [2]. The preclinical stages, when we can already detect these changes, but the subject does not have symptoms yet, constitute a valuable therapeutic window to administer disease-modifying drugs before irreversible neurodegenerative damage develops [2, 3]. However, cost-effective, non-invasive, and population-level useful biomarkers are needed to identify suitable patients.

In the last few years, the field of AD has undergone a revolution with the development of high-sensitivity assays that identify plasma markers such as Aβ₄₂/Aβ₄₀, phosphorylated tau at Thr181 (p-tau181), Thr217 (p-tau217), Thr231 (p-tau231), neurofilament light, and glial fibrillary acid protein [4 –9], opening up the possibility of avoiding other invasive, more expensive, and less available diagnostic methods such as lumbar puncture or positron emission tomography (PET) scans. Plasma markers have been shown to be highly accurate in distinguishing between AD pathology and other dementias [6, 10], and several studies suggest that the different p-tau species in plasma are sensitive markers of cerebral amyloid deposition [7].

With disease-modifying treatments on the verge of approval, it becomes necessary to understand these markers better, as they can help select suitable treatment candidates and monitor their response [11]. Moreover, they might provide us with accurate and reliable information in the preclinical stages of AD. In this regard, plasma markers offer valuable opportunities. Their accessibility makes it possible for sequential assessment, which may increase their diagnostic potential [12]. Thus, we can monitor changes in biomarker levels within a single subject rather than comparing them with the population. Thus, by understanding the evolution of plasma markers in the AD continuum, we could choose the earliest ones as a population screening tool in cognitively unimpaired (CU) subjects and also select the most appropriate markers to monitor the response to treatment.

However, these markers also present challenges, and many questions remain to be answered. It is unknown which marker is most useful at which stage of the disease, what time intervals are necessary to detect representative changes, and the relationship of these changes over time with neuropsychological or structural and functional imaging measures.

P-tau231 has emerged as one of the most promising plasma biomarkers. Previous studies in CSF and, more recently in plasma, suggest that p-tau231 alters very early, along with the first Aβ changes [7, 13]. Although its performance has been evaluated in patients with mild cognitive impairment and AD in cross-sectional studies, sequential changes of p-tau231 levels in the preclinical stages of AD need to be assessed. In this longitudinal study, we aim to provide information about the short-term evolution of p-tau231 in the preclinical stages of AD and its relationship with other parameters, such as neuropsychological or imaging markers.

MATERIALS AND METHODS

Participants

We performed this research with volunteers of the ‘Valdecilla Cohort for the study of memory and brain aging’ from the Memory Unit of the Marqués de Valdecilla University Hospital [14, 15]. The Valdecilla cohort is designed to help us better understand the preclinical phases of AD. It comprises CU volunteers who responded to an open call in the local media. Inclusion criteria are 1) age≥55 years; 2) signed consent for the collection and storage of biological samples and the performance of invasive techniques. Exclusion criteria are 1) cognitive impairment (determined by Clinical Dementia Rating (CDR)>0 [16]; 2) major psychiatric pathology; 3) significant systemic disease or sensory deprivation impairing the performance of the cognitive tests; 4) any contraindication to perform the complementary tests such as anticoagulation or claustrophobia.

At baseline, all participants were assessed with a comprehensive neuropsychological (NPS) battery conducted by neuropsychologists specialized in dementia, a brain magnetic resonance imaging (MRI), and an 18-fluorodeoxyglucose positron emission tomography (FDG-PET). DNA samples were extracted for genetic analyses, and CSF AD markers were determined (Aβ₄₂, Aβ₄₀, p-tau181, and t-tau). Annual follow-ups include plasma collection and a NPS assessment. One hundred and forty-six Caucasian participants were studied for baseline plasma p-tau231, one hundred and twenty-three had data from the first follow-up visit and sixteen for a second follow-up visit. All participants remained CU during the study.

This research has been conducted in accordance with the Declaration of Helsinki and has been approved by the ethics committee of the Hospital Universitario Marqués de Valdecilla. Title: Valdecilla Cohort for the study of memory and brain aging. Internal code: 2018.111. All subjects have given their signed consent to participate.

Plasma and CSF collection

Baseline plasma and CSF extractions were performed with the subjects fasting, on the same day, between 9 and 10 AM and a difference of less than 30 min between them. Our institution is part of the Alzheimer’s Association quality control program, so we follow international recommendations for plasma and CSF collection and storage [17, 18]. Aβ, p-tau, and t-tau values were determined using Fujirebio’s automated immunoassay analyzer Lumipulse G600 II [19] with the kits Lumipulse G β-Amyloid 1-40 (lot 4YX3085), Lumipulse G β-Amyloid 1-42 (lot 7ZX3084), Lumipulse G p-tau181 (lot 5DX3055) and Lumipulse G t-tau (lot 6BX3064).

Plasma samples were obtained following the standardized operating procedure described elsewhere [20]. In short, the blood is stored in 10 ml EDTA tubes and kept cold until processing within the next three hours. Then, it is centrifuged at 1800 g for 10 min. The supernatant is stored in polypropylene tubes in volumes of 500μl and frozen at -80°C until analyzed at the Clinical Neurochemistry Laboratory at the University of Gothenburg, Sweden. Plasma P-tau231 concentration was measured using an in-house ultrasensitive Single molecule array (Simoa) assay on an HD-X Analyzer (Quanterix, Billerica, MA, USA). Lower limit of quantification (2 pg/ml) and the rest of the analytical parameters can be consulted in previous studies [13].

CSF biomarkers

Participants were categorized according to ATN classification [21] based on their CSF biomarkers at baseline. As our cohort consists of CU subjects, instead of using the cut-off points established by the manufacturer in the kits mentioned above, they were established using an unbiased Gaussian mixture modelling [22]. We have dichotomized these variables and considered Aβ-positive (A+) when CSF Aβ_42/40 ratio < 0.076, tau-positive (T+) when CSF p-tau181 > 73.2 pg/ml, and neurodegeneration-positive (N+) if CSF t-tau>543 pg/ml.

Magnetic resonance and PET imaging

All MRI scans were performed on the same 3T Philips Medical Systems MRI scanner with an 8-channel head coil. To determine the volume of the different structures, a sagittal T1-weighted MPRAGE sequence was used (170 slices, 1.2 mm voxel size, 9° of flip angle and shortest echo and repetition times) with subsequent processing as described in previous publications [14]. To segment the hippocampus, we used the automated FreeSurfer protocol, version 6.0 (http://surfer.nmr.mgh.harvard.edu). This protocol includes skull stripping, labeling of the volumes of each segmentation, and normalization of voxel intensities. Cortical and subcortical volume measurements were derived using the surface and subcortical segmentation pipelines [23]. Subcortical measurements were automatically derived from the subcortical processing flow. The quality check was performed using the ENIGMA Consortium quality control protocol (http://enigma.ini.usc.edu/).

The synthesis of ¹⁸F-FDG and the acquisition of PET images were carried out in our hospital’s Department of Nuclear Medicine. ¹⁸F was obtained using the 18O(p,n)18F nuclear reaction in a cyclotron PET Trace (General Electric Healthcare, Wisconsin, USA) and ¹⁸F-FDG scans were acquired in a Siemens Biograph LSO Pico 3D equipment (Siemens Healthcare Molecular Imaging, IL, USA). Subjects were administered intravenously with 3–4 MBq/kg ¹⁸F-FDG. Image acquisition consisted of a static image acquired 30-45 min after tracer injection. Images were reconstructed on a 128 x 128 matrix using the ordered subsets expectation maximization iterative method (six iterations and 16 subsets, zoom 2.5, filter FWHM 2 mm and sinogram trim factor 1.5). The axial slices were reoriented parallel to the frontal-occipital axis. The semiquantitative analysis of each study was performed using the Scenium ^® program integrated into the Siemens Syngo.via Neurology software (Siemens CTI Molecular Imaging, Knoxville, TN, USA). The average standardized uptake value (SUV) ratio values in regions of interest (ROIs) from the Automated Anatomical Labelling Atlas [24] were used. We have taken the cerebellum as a reference.

APOE testing

Apolipoprotein E (APOE) genotype was determined by TaqMan SNP genotyping assays (Applied Biosystems, Foster City, CA, United States). Subjects carrying≥1 copy of the ɛ4 allele were considered ɛ4 + and the rest were considered ɛ4-.

Neuropsychological tests

Participants in our cohort performed a comprehensive neuropsychological battery designed to detect early signs of AD. It includes a global assessment through the CDR and the Mini-Mental State Examination (MMSE) [25], and several tests to evaluate the different cognitive domains. For memory assessment, we used the validated Spanish version of the Free and Cued Selective Reminding Test (FCSRT) [26], a test designed to measure verbal learning and memory using a list learning. We have considered different items such as the total free recall (TFR), total recall (TR), delayed free recall (DFR), and delayed total recall (DTR). For narrative memory, we used the logical memory (LM) subtest of the Weschler Memory Scale-III [27], which is based on the telling of two stories that are recalled in an immediate and delayed form (DR). In this test, we considered the total delayed units (LMDU). We also performed the Spanish version of the Face Name Associative Memory Exam (FNAME), a highly demanding face-name associative memory exam [28]. It consists of asking subjects to recall sixteen faces associated with a name and an occupation each. The first section includes initial recall of face-occupation and face-name pairs, followed by immediate recall of the name and occupation associated with each face. After 30 min, the subject is asked to recall as much of the information associated with each face as possible. The total value (TFNAME) is the number of all names and occupations recalled throughout the test and it is the measure we have considered for this study. The executive functions were assessed by means of the two parts of the Trail Making Test (TMT-A and TMT-B) [29] and the Digit-Symbol Substitution Test [26]. The Preclinical Alzheimer’s cognitive composite (PACC) [30] was also constructed with the results of the NPS assessment (we have used the PACC5 version, which considers the MMSE, the DR of the Logical Memory subtest, the Digit-Symbol test, the DFR and DTR of the FCSRT, and categorical fluency).

Statistical analysis

The assessment of normality was conducted using Kolmogorov-Smirnov test, and Shapiro-Wilk test. We used Student’s t-test for paired samples to analyze the differences between p-tau231 concentrations at baseline and follow-up and provided 95% confidence intervals (CI). We also used Student’s test-test for paired samples after stratifying subjects according to their ATN group followed by the Bonferroni correction to adjust p-values (p-values presented in the text are adjusted p-values). Additionally, we performed the analysis of variance (ANOVA) followed by Tukey’s test to examine the differences between the ATN groups. Given the small sample size of the second follow-up visit (N = 16), for this analysis we have only considered the baseline visit and the first follow-up. The effect size of the mean analysis was analyzed using Cohen’s d.

We performed a Linear Mixed-effects Model (LMM) to analyze the rate of change of the p-tau231 levels. LMM are useful when we have data with more than one source of random variability. The repeated measures taken over time allow to account for both within-person and across-person variability. The models performed considered the plasma p-tau231 concentrations as the dependent variable and number of days between plasma collections as the time variable. A random intercept and slope model was performed, both conditions allow everyone to have a different level of p-tau231 at baseline and different slope over time. Also, the ATN group was considered as an explanatory variable and age, sex, and APOE ɛ4 status as covariates. Furthermore, p-tau231 levels were skewed and transformed (tfp-tau231) by operating the Box-Cox transformation to normalize the distributions. We have used this technique to improve the accuracy of the predictions made by LMM since it stabilizes the variance and makes it easier to interpret the data.

As p-tau231 levels did not follow a normal distribution, we used Spearman’s Rho to study the correlation between the plasma biomarker level and CSF, neuropsychological and imaging markers. Thus, we correlated baseline p-tau231 values with baseline CSF biomarkers (Aβ₄₀, Aβ₄₂, Aβ₄₂/Aβ₄₀ ratio, p-tau181, and t-tau), MRI measures (right and left hippocampal volume), PET-FDG (regional metabolism in right and left hippocampus, right and left precuneus, and globally) and with the main outputs of FCSRT, FNAME, LM, and TMT test. We also correlated baseline p-tau231 with cognitive tests stratified by the ATN group. Afterward, we selected the significant results and adjusted for covariates using a univariate generalized linear model with cognitive measures as the dependent variable. Missing data have been handled by case removal.

Statistical analyses were performed using SPSS Statistics V.20.0 (IBM, NY, USA) and R Statistical Software (version 4.3.0; R Foundation for Statistical Computing, Vienna, Austria).

RESULTS

Sample description

The sample includes 146 cognitively unimpaired subjects (average MMSE score of 28.8±1.5 and average DTR of 14.7±1.95), with a mean age of 64.7±8.33 years, and female predominance (71.9%). Ninety-one participants were A-T-N- (74.7% females), 40 A + T-N- (70% females), and 15 A + T+N+(60% females). The mean age was significantly higher in the A + T-N- group than the A-T-N- group (66.8 and 63.7 years, respectively; p = 0.007) and in the A + T+N+group than the A-T-N- (70.1 and 63.7 years respectively; p = 0.0004). However, it was not significantly different between the A + T-N- and A + T+N+subjects (66.8 and 70.1 years; p = 0.06). The proportion of APOE ɛ4 carriers was 29.4%, and 37.7% of subjects had at least one positive AD marker in the CSF (Table 1). The average time between the collection of the baseline plasma samples and the first follow-up visit was 267.3±86.5 days, and between the first and second follow-ups was 343.7±62.1 days.

Table 1

Baseline characteristics of the subjects

Characteristic	N = 146
Females, n (%)	105 (71.9%)
Age, mean (SD)	64.7 (8.3)
APOEɛ4 carrier, n. (%)	43 (29.4%)
MMSE (0–30), mean (SD)	28.8 (1.5)
FCSRT delayed total recall (0–16), mean (SD)	14.8 (2.4)
CSF Biomarkers
Aβ₄₀, mean (SD), pg/ml	10,577.6 (3256.5)
Aβ₄₂, mean (SD), pg/ml	796.1 (327.4)
Ratio Aβ_42/40, mean (SD)	0.076 (0.02)
Total-tau, mean (SD), pg/ml	336.4 (147.3)
Phosphorylated tau, mean (SD), pg/ml	45.2 (28.1)
ATN group, n (%)
A-T-N-	91 (62.3%)	Female: 68 (74.7%)
		Male: 23 (25.3%)
		Age, mean (SD): 63.7 (6.2)
A+T-N-	40 (27.4%)	Female: 28 (70.0%)
		Male: 12 (30.0%)
		Age, mean (SD): 66.8 (5.8)
A+T+N+	15 (28.1%)	Female: 9 (60.0%)
		Male: 6 (40.0%)
		Age, mean (SD): 70.1 (5.2)

n, number of subjects; SD, standard deviation; MMSE, Mini-Mental State Examination; FCSRT, Free and Cued Selective Reminding Test; CSF, cerebrospinal fluid; Aβ, amyloid-beta; A, amyloid; T, tau; N, neurodegeneration.

Longitudinal analysis

Plasma p-tau231 was analyzed in 146 subjects at baseline, 123 at first follow-up, and 16 individuals at second determination. Correlations between baseline plasma p-tau231 and CSF biomarkers can be found in Supplementary Table 1. The changing of the p-tau231 concentrations in our sample was examined in two ways. First, we tested whether the average of p-tau231 was higher as the visits progressed (only at the baseline and first follow-up, due to the small sample size of the second follow-up); and second, we analyzed the rate of change of the biomarkers over time (days elapsed). The former analysis was performed to test the hypotheses that the sample experienced a higher average of p-tau231 in each collection, while the latter allows us to estimate the rate of change of p-tau231 concentrations per day elapsed.

Plasma p-tau231 differences between timepoints

The baseline p-tau231 values deviated from normality (Shapiro Wilk’s test p = 0.0003) and so did the levels at first follow-up (Shapiro Wilk’s test p = 0.006) and the change in p-tau231 from baseline to first follow-up (Shapiro Wilk’s test p = 0.0003). The mean baseline p-tau231 was 11.4 pg/ml±5.98 and 13.3 pg/ml±6.04 at the first follow-up. The difference between baseline and first follow-up levels was statistically significant (1.99 pg/ml; 95% CI 1.22 to 2.77; p < 0.0001; Cohen’s d = 0.36) (Fig. 1).

Fig. 1

Longitudinal changes in p-tau231. The Y axis represents plasma p-tau231 concentration expressed in pg/ml and the X axis, the sequential visits. In the boxplot, the boxes show the interquartile range (the upper boundary is the Q3, and the lower boundary is the Q1). The line inside the box corresponds to the median of the sample and the whiskers represent the maximum (upper) and minimum (lower) values. 146 subjects were studied at baseline and 123 at first follow-up. The mean baseline was 11.40 pg/ml±5.98, and at first follow-up it was 13.28 pg/ml±6.04. The mean difference between baseline and first visit was significant (p = 0.0000013).

At baseline, the different ATN groups showed significantly different values of p-tau231 (p < 0.001). The difference between the A-T-N- and A + T-N- groups was not significant (1.88 pg/ml; p = 0.19); but it was significant between the A-T-N- and A + T+N+groups (6.27 pg/ml; p = 0.0003; Cohen’s d = 1.11) and between the A + T-N- and A + T+N+groups (4.38 pg/ml; p = 0.03; Cohen’s d = 0.73). The same happened at the first follow-up visit, with significant differences between the different ATN groups (p = 0.0001). Between the A-T-N- and A + T-N- groups there were no significant differences (1.74 pg/ml; p = 0.26) but there were among the A-T-N- and A + T+N+subjects (7.32 pg/ml; p < 0.0001; Cohen’s d = 1.47) and between A + T-N- and A + T+N+(5.58 pg/ml; p = 0.007; Cohen’s d = 0.77).

When we have assessed the differences between time points and stratified by ATN group, in the A-T-N-, the mean baseline p-tau231 was 10.1 pg/ml (±5.00), and at first follow-up visit it was 11.9 pg/ml (±4.51). In this group, the mean difference was 1.79 pg/ml (95% CI = 2.82 to 0.76; p = 0.03; Cohen’s d = 0.35). In the A + T-N- group, the mean baseline p-tau231 value was 11.8 pg/ml (±5.24) and 14.2 pg/ml (±6.4) at follow-up. The difference was also significant in this group, with a value of 2.36 pg/ml (95% CI = 3.56 to 1.16; p = 0.001; Cohen’s d = 0.37). Finally, in the A + T+N+group, the mean value of p-tau231 at baseline was 17.4 pg/ml (±8.0) and at follow-up 19.6 pg/ml (±8.88). The average difference in this group was not significant, with a value of 2.24 pg/ml (95% CI = 5.83 to 1.34; p = 0.58; Cohen’s d = 0.46) (Fig. 2). The p-tau231 values per time point and ATN group and the differences between them can be found in Supplementary Tables 2 and 3, respectively.

Fig. 2

Changes in p-tau231 stratified by ATN group. The Y axis represents plasma p-tau231 concentration expressed in pg/ml and the X axis, the baseline and first visit. The boxes show the interquartile range (the upper boundary is the Q3, and the lower boundary is the Q1). The line inside the box corresponds to the median of the sample and the whiskers represent the maximum (upper) and minimum (lower) values. The dots outside the boxes indicate outliers. White boxes correspond to A-T-N- group, grey boxes to A + T-N- and the black ones to the A + T+N+group. An increase in p-tau231 values is observed between visits. The mean differences in the A-T-N- and A + T+N+groups were significant in both baseline and first visit (p = 0.0003 and p < 0.0001 respectively; asterisks with lines). The same happened between the A + T-N- and A + T+N+groups (p = 0.03 at baseline and p = 0.007 in the first visit; asterisks).

Longitudinal trajectory of Plasma p-tau231 concentrations over time

We analyzed the longitudinal change of tfp-tau231 concentrations considering the influence of variables such as age, sex, presence of APOE ɛ4 and ATN group. We estimated a tfp-tau231 increase of 0.0025 pg/ml/day (p < 0.0001). Sex did not significantly influence the tfp-tau231 basal levels, nor did the age (Estimate=-0.46 pg/ml, p = 0.22 and Estimate = 0.05 pg/ml, p = 0.08, respectively). APOE ɛ4 carriers showed a predicted increase in baseline tfp-tau231 levels of 0.8 pg/ml over non-carriers (p = 0.03), and the A + T + N + group presented an estimated baseline 1.62 pg/ml higher than the A-T-N- group (p = 0.01). However, the ATN and ApoE groups showed no interactions with the growth rate of tfp-tau231 over time (Fig. 3). Standardized estimates can be found in Supplementary Table 3.

Fig. 3

Longitudinal trajectory of plasma p-tau231 concentrations over time (days) by ATN group, segmented by APOE, and adjusted by sex and age at baseline. In this linear mixed model plot the dots represent each individual value. The shaded area around the regression lines (continuous lines) represents the 95% confidence interval. The Y axis represents the Box-cox transformed plasma p-tau231 values. Dashed lines represent the individual trajectories of each subject. A-T-N- subjects are shown in red, A + T-N- in green, and A + T+N+in blue. In the X axis we show number of days between samples. We divided the plot between non-carriers of the ApoE4 allele (left side) and carriers of at least one copy (right side). There are no differences between slopes, but there are differences in the general level. The A + T+N+group showed to have an estimated baseline 1.62 units higher than the A-T-N- group (p = 0.014). The ATN groups showed no interactions with the different follow-up visits in the growth rate of p-tau231.

Correlations with neuropsychological and imaging measurements

We correlated p-tau231 plasma levels and their rate of change between baseline and the first follow-up visit with a wide range of neuropsychological tests and imaging measurements (Table 2).

Table 2

Correlation between plasma p-tau231 and both neuropsychological and imaging measurements

				Baseline plasma p-tau231 (pg/ml)	Plasma p-tau231 change between baseline and first follow-up (pg/ml)
		N		Spearman’s Correlation coefficient	Spearman’s Correlation coefficient
NPS	LMDU	146	Rho	–0.16	–0.15
			p	0.057	0.09
	DFR	146	Rho	–0.12	–0.09
			p	0.14	0.33
	DTR	144	Rho	–0.25	–0.03
			p	0.002	0.77
	TMT-A	146	Rho	0.16	0.07
			p	0.051	0.41
	TMT-B	139	Rho	0.10	0.07
			p	0.22	0.43
	TFNAME	146	Rho	–0.25	0.03
			p	0.002	0.76
	PACC5	146	Rho	–0.08	–0.03
			p	0.35	0.76
MRI	Left hippocampus	107	Rho	0.12	0.01
			p	0.20	0.94
	Right hippocampus	107	Rho	–0.02	0.01
			p	0.85	0.88
FDG PET	Average	136	Rho	–0.05	–0.02
			p	0.59	0.84
	Right precuneus	136	Rho	–0.03	–0.07
			p	0.68	0.42
	Left precuneus	136	Rho	–0.03	–0.04
			p	0.76	0.66
	Right hippocampus	136	Rho	–0.09	0.06
			p	0.31	0.52
	Left hippocampus	136	Rho	–0.12	0.08
			p	0.17	0.36

NPS, neuropsychological; MRI, magnetic resonance imaging; FDG PET, fluorodeoxyglucose positron emission tomography; N, number of subjects; p, statistical significance; Rho, correlation coefficient; LMDU, total delayed units in Logical Memory test; DFR, delayed free recall in Free and Cued Selective Reminding test; DTR, Delayed Total Recall; TMT-A, Trail Making Test part A; TMT-B, Trail Making Test part B; TFNAME, total value of Face Name Associative Memory Exam; PACC5, preclinical Alzheimer cognitive composite.

P-tau231 levels at baseline showed no correlation with the total right (r=-0.01; p = 0.85) or left (r=-0.12; p = 0.23) hippocampal volume. Nor did the change in p-tau231 over time correlate significantly with left (r = 0.01; p = 0.94) or right (r = 0.01; p = 0.88) hippocampal volume. Considering the effect of age, the uptake of FDG-PET did not correlate significantly with baseline p-tau231 values in different regions of interest, such as right precuneus (r=-0.03; p = 0.68), left precuneus (r=-0.03; p = 0.76), right hippocampus (r=-0.09; p = 0.31), left hippocampus (r=-0.12; p = 0.17) and neither with global average uptake (r=-0.05; p = 0.59).

However, baseline p-tau231 levels showed a correlation with different neuropsychological tests, such as the DTR of the FCSRT (r=-0.25; p = 0.002) and with the TFNAME (r=-0.25; p = 0.002). These tests did not correlate with the change of p-tau231 over time (r=-0.026; p = 0.77 and r = 0.03; p = 0.76 respectively). Neither PACC5 was significantly correlated with baseline p-tau231 levels (r=-0.08; p = 0.35) nor with difference between samples (r=-0.03; p = 0.76).

To better understand our results, we stratified the baseline correlations of p-tau231 with neuropsychological tests by ATN groups. P-tau231 levels did not correlate significantly with TFNAME in the A-T-N- group (r=-0.03; p = 0.75), but it did in the amyloid-positive groups, in which the magnitude of the correlation increased progressively: A + T-N- (r=-0.37; p = 0.02) and A + T+N+(r=-0.67; p = 0.01) (Table 3). We further examined the association between baseline p-tau231 and TFNAME by ATN group accounting for age, sex, and APOE4 status by a generalized linear model. Using p-tau231 as a dependent variant, we found that in the A + T-N- group, the relationship with TFNAME remained significant (Estimate=-4.2; p = 0.04) after adjusting for age, sex, and APOE. This did not happen in the A + T+N+group (Estimate=-9.1; p = 0.16), though the number of individuals included in this analysis was limited (N = 14).

Table 3

Correlation between baseline plasma p-tau231 and cognitive tests stratified by ATN group

	Cognitive test	A-T-N-		A+T-N-		A+T+N+
	Rho	p	Rho	p	Rho	p
Plasma p-tau231	LMDU	0.01	0.88	–0.16	0.32	–0.33	0.25
	DFR	–0.03	0.77	–0.04	0.8	–0.43	0.12
	DTR	–0.12	0.27	–0.18	0.27	–0.43	0.12
	TMT-A	0.02	0.86	0.02	0.9	0.47	0.08
	TMT-B	–0.05	0.67	0.01	0.96	0.53	0.75
	TFNAME	–0.03	0.75	–0.37	0.02	–0.67	0.01
	PACC5	–0.11	0.29	–0.22	0.18	–0.002	0.9

p, statistical significance; Rho, correlation coefficient; LMDU, total delayed units in Logical Memory test; DFR, delayed free recall in Free and Cued Selective Reminding test; DTR, Delayed Total Recall; TMT-A, Trail Making Test part A; TMT-B, Trail Making Test part B; TFNAME, total value of Face Name Associative Memory Exam; PACC5, preclinical Alzheimer cognitive composite; A, amyloid; T, tau; N, neurodegeneration.

PACC5 did not show significant correlations with baseline p-tau231 in A-T-N- subjects (r=-0.1; p = 0.3), A + T-N- (r=-0.2; p = 0.18) nor in the A + T+N+group (r=-0.002; p = 0.9).

DISCUSSION

Novel plasma biomarkers for AD are a promising tool for diagnosing and classifying patients in hospital and primary care settings. Among them, p-tau231 rises in plasma very early and allows the detection of subjects at risk for AD [7, 13]. Previous studies in cognitively unimpaired individuals modelling the trajectory of p-tau231 showed that it significantly increased after amyloid burden passed the threshold of 26.4 centiloids on Aβ PET [7]. In the same study, p-tau217 levels also raised very early but required a higher threshold of amyloid pathology. On the other hand, p-tau181 is a very specific marker of AD and predicts cognitive impairment and hippocampal atrophy within one year, but it only increases significantly in plasma with much higher amyloid loads [7, 31]. The results of our study show that plasma p-tau231 levels progressively increase in CU subjects with a mean age of 64.8 years, in samples obtained with one year of difference on average. This change is appreciable even in A-T-N- subjects who would be considered as not in the AD continuum.

Although it could constitute an ultra-early sign of amyloid dysmetabolism, its significance is still unclear and does not correlate with any cognitive, functional, or structural trait. To clarify this aspect of our findings, longer-term studies determining p-tau231 in A-T-N- subjects are necessary. Evaluating the evolution of p-tau231 in subjects who convert from A- to A + during follow-up would be a relevant issue. We have been unable to assess this point because we do not perform serial CSF measurements in our cohort. A similar trend of increase of p-tau231 from baseline to the follow-up visit in A + T-N- and A + T+N+was found, though this was only significant in A + T-N-, which could be explained by the low numbers of the A + T+N+group (N = 12). This is a limitation of our study; therefore, we might be underpowered for some analysis.

The linear mixed model estimated that the growth per day was 0.0025 units of tfp-tau231, and we did not find an interaction with the ATN groups; therefore, the slopes of the increase we observed in the three ATN groups were not significantly different. This would point toward a linear increase in our population of subjects around 65 years of age on average independent of their ATN status. However, as mentioned above, some of our subgroups (especially the A + T+N+) had a very low number of participants, and we might be underpowered to detect interactions. In this sense, although subjects showed significantly older ages in the progressive ATN groups, age did not show an effect on baseline p-tau231 levels in our LMM, but the A + T+N+group did show a baseline increase of 1.62 units compared to the A-T-N- (and a difference of 6.27 pg/ml in the study of means), pointing out that plasma p-tau231 levels might be more related to AD pathology than to age. However, the standard deviation in the A + T+N+group was 8.01. This fact makes it difficult to interpret the changes in p-tau231 along the AD continuum, thus requiring longer-term studies and larger sample sizes to confirm which periods are necessary to identify changes in the marker greater than its intrinsic variability.

A relevant aspect of our cohort is that it encompasses the age spectrum in which AD is known to begin to increase in incidence [32]. Even though age was not significantly associated with p-tau231, we speculate that our population might be suitable to detect early p-tau231 levels increase, reflecting an early stage in the neurodegeneration process.

Longitudinal studies with post-mortem anatomopathological examination show that p-tau231 levels also increase in the late stages of AD and are higher than those of subjects with mild amyloid pathology [33]. This gradual correlation with the main pathological signatures of AD in clinicopathological studies suggests that the p-tau231 increase might be related to early AD changes. In this sense, p-tau231 has shown to be able to discriminate between Braak 0 and Braak I–II stages [34]. However, recent evidence indicates that, although p-tau217 increases progressively along the AD continuum, p-tau231 reaches a plateau and may not be useful for monitoring disease progression or long-term treatments efficacy [32]. P-tau181 appears to act similarly to p-tau217, as it increases progressively along the AD continuum, including the preclinical stages, and its levels correlate well with amyloid pathology [35, 36].

To characterize better the early preclinical phase of AD, we correlated p-tau231 levels with neuropsychological tests and measures of brain volume and metabolism. We found that the change from baseline to the follow-up did not correlate with any of these measures. However, baseline levels were associated with cognitive measurements. In contrast to what appears to happen with p-tau217 and p-tau181, which are associated with cortical atrophy in CU subjects [37, 38], we did not detect any significant correlation with structural measures such as hippocampal volume or glucose metabolism of different regions of interest for AD.

However, baseline levels of p-tau231 correlated with poorer cognitive outcomes, measured by sensitive neuropsychological memory tests such as the FNAME and FCSRT. Interestingly, when we stratified by ATN, this happened only from the A + T-N- group onwards and progressively increased along the AD continuum, and this correlation remained significant after adjusting for covariates. These findings are in line with another previous study that has shown a correlation between baseline p-tau231 levels and longitudinal changes in cognition, determined by the PACC at three years, only in Aβ+subjects [7]. Our interpretation of this finding is that we can only detect an association between tau pathology (plasma levels of p-tau231) and cognitive impairment in individuals with significant amyloid deposition (positive Aβ_42/40 ratio in CSF).

For its part, in more recent studies, baseline levels of plasma p-tau217 have been shown to correlate with a worsening of memory (word list delayed recall) in a group of CU presenilin-1 E280A carriers [39] and increases in this biomarker also correlate with a worsening in modified PACC and MMSE over as long as six years in Aβ+CU subjects [35]. P-tau181 is not altered as early as p-tau217 and p-tau231 but has shown to be a good predictor of progression to AD and is longitudinally related to cognitive decline through parameters such as the MMSE, CDR and PACC [40].

Some limitations of our study should be pointed out. First, we only have CSF from the baseline visit, so we cannot confirm whether any of the subjects changed their ATN group during the follow-up time. Furthermore, although the changes in p-tau231 between timepoints are significant, the effect size is small, and this may difficult its application in clinical practice. On the other hand, even though our main findings are statistically sound, as we have mentioned above, the small sample size is a limitation for some of the stratified analyses so our results should be taken with caution and replications, ideally in larger and more ethnically heterogeneous samples, would be needed to confirm our conclusions.

Conclusions

Our results suggest that time intervals as short as one year could be meaningful in detecting significant changes in plasma p-tau231 values at a group level in CU participants, even in A-T-N- subjects. Sequential testing of plasma biomarkers could play an important role when selecting people for clinical trials and, in the near future, for initiating disease-modifying treatments in an efficient and non-invasive way. Still, long-term longitudinal studies are needed to give us a better understanding of its temporal evolution.

AUTHOR CONTRIBUTIONS

Francisco Martínez Dubarbie (Conceptualization; Formal analysis; Investigation; Writing – original draft); Sara López-García (Investigation; Writing – review & editing); Carmen Lage (Investigation; Writing – review & editing); Guglielmo Di Molfetta (Investigation; Writing – review & editing); Marta Fernández-Matarrubia (Investigation; Writing – review & editing); Ana Pozueta-Cantudo (Investigation; Writing – review & editing); María García-Martínez (Investigation; Writing – review & editing); Andrea Corrales-Pardo (Investigation; Writing – review & editing); María Bravo (Investigation; Writing – review & editing); Julio Jiménez-Bonilla (Investigation; Writing – review & editing); Remedios Quirce (Investigation; Writing – review & editing); Enrique Marco de Lucas (Investigation; Writing – review & editing); Marta Drake-Pérez (Investigation; Writing – review & editing); Diana Tordesillas (Investigation; Writing – review & editing); Marcos López-Hoyos (Investigation; Writing – review & editing); Juan Irure-Ventura (Investigation; Writing – review & editing); Elizabeth Valeriano-Lorenzo (Formal analysis; Software; Writing – review & editing); Kaj Blennow (Investigation; Resources; Writing – review & editing); Nicholas J. Ashton (Investigation; Resources; Writing – review & editing); Henrik Zetterberg (Investigation; Resources; Writing – review & editing); Eloy Rodríguez-Rodríguez (Conceptualization; Investigation; Methodology; Supervision; Writing – original draft); Pascual Sánchez-Juan (Conceptualization; Formal analysis; Investigation; Methodology; Supervision; Writing – original draft).

Footnotes

ACKNOWLEDGMENTS

We would like to thank all the Valdecilla Cohort volunteers for their participation.

FUNDING

KB is supported by the Swedish Research Council (#2017-00915 and #2022-00732), the Swedish Alzheimer Foundation (#AF-930351, #AF-939721 and #AF-968270), Hjärnfonden, Sweden (#FO2017-0243 and #ALZ2022-0006), the Swedish state under the agreement between the Swedish government and the County Councils, the ALF-agreement (#ALFGBG-715986 and #ALFGBG-965240), the European Union Joint Program for Neurodegenerative Disorders (JPND2019-466-236), the Alzheimer’s Association 2021 Zenith Award (ZEN-21-848495), and the Alzheimer’s Association 2022-2025 Grant (SG-23-1038904 QC).

CONFLICT OF INTEREST

KB has served as a consultant and at advisory boards for Acumen, ALZPath, BioArctic, Biogen, Eisai, Julius Clinical, Lilly, Novartis, Ono Pharma, Prothena, Roche Diagnostics, and Siemens Healthineers; has served at data monitoring committees for Julius Clinical and Novartis; has given lectures, produced educational materials and participated in educational programs for Biogen, Eisai and Roche Diagnostics; and is a co-founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program, outside the work presented in this paper.

Henrik Zetterberg is an Editorial Board Member of this journal but was not involved in the peer-review process of this article nor had access to any information regarding its peer-review.

All other authors have no conflict of interest to report.

DATA AVAILABILITY

The data supporting the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

The supplementary material is available in the electronic version of this article: .

References

Jack

Jr , Knopman

, Jagust

, Shaw

, Aisen

, Weiner

, Petersen

, Trojanowski

(2010) Hypothetical model of dynamic biomarkers of the Alzheimer’s pathological cascade. Lancet Neurol 9, 119–128.

Jack

Jr , Bennett

, Blennow

, Carrillo

, Dunn

, Haeberlein

, Holtzman

, Jagust

, Jessen

, Karlawish

, Liu

, Molinuevo

, Montine

, Phelps

, Rankin

, Rowe

, Scheltens

, Siemers

, Snyder

, Sperling R; Contributors (2018) NIA-AA Research Framework: Toward a biological definition of Alzheimer’s disease. Alzheimers Dement 14, 535–562.

Cummings

, Lee

, Zhong

, Fonseca

, Taghva

(2021) Alzheimer’s disease drug development pipeline: 2021. Alzheimers Dement (N Y) 7, e12179.

Simrén

, Leuzy

, Karikari

, Hye

, Benedet

, Lantero-Rodriguez

, Mattsson-Carlgren

, Schöll

, Mecocci

, Vellas

, Tsolaki

, Kloszewska

, Soininen

, Lovestone

, Aarsland

; AddNeuroMed consortium; Hansson

, Rosa-Neto

, Westman

, Blennow

, Zetterberg

, Ashton

(2021) The diagnostic and prognostic capabilities of plasma biomarkers in Alzheimer’s disease. Alzheimers Dement 17, 1145–1156.

Benedet

, Milà-Alomà

, Vrillon

, Ashton

, Pascoal

, Lussier

, Karikari

, Hourregue

, Cognat

, Dumurgier

, Stevenson

, Rahmouni

, Pallen

, Poltronetti

, Salvadó

, Shekari

, Operto

, Gispert

, Minguillon

, Fauria

, Kollmorgen

, Suridjan

, Zimmer

, Zetterberg

, Molinuevo

, Paquet

, Rosa-Neto

, Blennow

, Suárez-Calvet

; Translational Biomarkers in Aging and Dementia (TRIAD) study, Alzheimer’s and Families (ALFA) study, and BioCogBank Paris Lariboisière cohort (2021) Differences between plasma and cerebrospinal fluid glial fibrillary acidic protein levels across the Alzheimer disease continuum. JAMA Neurol 78, 1471–1483.

Palmqvist

, Janelidze

, Quiroz

, Zetterberg

, Lopera

, Stomrud

, Su

, Chen

, Serrano

, Leuzy

, Mattsson-Carlgren

, Strandberg

, Smith

, Villegas

, Sepulveda-Falla

, Chai

, Proctor

, Beach

, Blennow

, Dage

, Reiman

, Hansson

(2020) Discriminative accuracy of plasma phospho-tau217 for Alzheimer disease vs other neurodegenerative disorders. JAMA 324, 772–781.

Milà-Alomà

, Ashton

, Shekari

, Salvadó

, Ortiz-Romero

, Montoliu-Gaya

, Benedet

, Karikari

, Lantero-Rodriguez

, Vanmechelen

, Day

, González-Escalante

, Sánchez-Benavides

, Minguillon

, Fauria

, Molinuevo

, Dage

, Zetterberg

, Gispert

, Suárez-Calvet

, Blennow

(2022) Plasma p-tau231 and p-tau217 as state markers of amyloid-β pathology in preclinical Alzheimer’s disease. Nat Med 28, 1797–1801.

Elahi

, Casaletto

, La Joie

, Walters

, Harvey

, Wolf

, Edwards

, Rivera-Contreras

, Karydas

, Cobigo

, Rosen

, DeCarli

, Miller

, Rabinovici

, Kramer

(2020) Plasma biomarkers of astrocytic and neuronal dysfunction in early- and late-onset Alzheimer’s disease. Alzheimers Dement 16, 681–695.

Ashton

, Janelidze

, Al Khleifat

, Leuzy

, van der Ende

, Karikari

, Benedet

, Pascoal

, Lleó

, Parnetti

, Galimberti

, Bonanni

, Pilotto

, Padovani

, Lycke

, Novakova

, Axelsson

, Velayudhan

, Rabinovici

, Miller

, Pariante

, Nikkheslat

, Resnick

, Thambisetty

, Schöll

, Fernández-Eulate

, Gil-Bea

, López de Munain

, Al-Chalabi

, Rosa-Neto

, Strydom

, Svenningsson

, Stomrud

, Santillo

, Aarsland

, van Swieten

, Palmqvist

, Zetterberg

, Blennow

, Hye

, Hansson

(2021) A multicentre validation study of the diagnostic value of plasma neurofilament light. Nat Commun 12, 3400.

10.

Thijssen

, La Joie

, Strom

, Fonseca

, Iaccarino

, Wolf

, Spina

, Allen

, Cobigo

, Heuer

, VandeVrede

, Proctor

, Lago

, Baker

, Sivasankaran

, Kieloch

, Kinhikar

, Yu

, Valentin

, Jeromin

, Zetterberg

, Hansson

, Mattsson-Carlgren

, Graham

, Blennow

, Kramer

, Grinberg

, Seeley

, Rosen

, Boeve

, Miller

, Teunissen

, Rabinovici

, Rojas

, Dage

, Boxer

; Advancing Research and Treatment for Frontotemporal Lobar Degeneration investigators (2021) Plasma phosphorylated tau 217 and phosphorylated tau 181 as biomarkers in Alzheimer’s disease and frontotemporal lobar degeneration: A retrospective diagnostic performance study. Lancet Neurol 20, 739–752.

11.

Ossenkoppele

, van der Kant

, Hansson

(2022) Tau biomarkers in Alzheimer’s disease: Towards implementation in clinical practice and trials. Lancet Neurol 21, 726–734.

12.

Sánchez-Juan

, Seshadri

(2017) Dynamic measurements of β-amyloid accumulation: The early effect of APOE. Neurology 89, 986–987.

13.

Ashton

, Pascoal

, Karikari

, Benedet

, Lantero-Rodriguez

, Brinkmalm

, Snellman

, Schöll

, Troakes

, Hye

, Gauthier

, Vanmechelen

, Zetterberg

, Rosa-Neto

, Blennow

(2021) Plasma p-tau231: A new biomarker for incipient Alzheimer’s disease pathology. Acta Neuropathol 141, 709–724.

14.

López-de-Eguileta

, López-García

, Lage

, Pozueta

, García-Martínez

, Kazimierczak

, Bravo

, Irure

, López-Hoyos

, Muñoz-Cacho

, Rodríguez-Perez

, Tordesillas-Gutiérrez

, Goikoetxea

, Nebot

, Rodríguez-Rodríguez

, Casado

, Sánchez-Juan

(2022) The retinal ganglion cell layer reflects neurodegenerative changes in cognitively unimpaired individuals. Alzheimers Res Ther 14, 57.

15.

López-García

, Lage

, Pozueta

, García-Martínez

, Kazimierczak

, Fernández-Rodríguez

, Bravo

, Reyes-González

, Irure

, López-Hoyos

, Rodríguez-Rodríguez

, Sánchez-Juan

(2021) Sleep time estimated by an actigraphy watch correlates with CSF tau in cognitively unimpaired elders: The modulatory role of APOE. Front Aging Neurosci 13, 663446.

16.

Morris

(1993) The Clinical Dementia Rating (CDR): Current version and scoring rules. Neurology 43, 2412–2414.

17.

Mattsson

, Andreasson

, Persson

, Arai

, Batish

, Bernardini

, Bocchio-Chiavetto

, Blankenstein

, Carrillo

, Chalbot

, Coart

, Chiasserini

, Cutler

, Dahlfors

, Duller

, Fagan

, Forlenza

, Frisoni

, Galasko

, Galimberti

, Hampel

, Handberg

, Heneka

, Herskovits

, Herukka

, Holtzman

, Humpel

, Hyman

, Iqbal

, Jucker

, Kaeser

, Kaiser

, Kapaki

, Kidd

, Klivenyi

, Knudsen

, Kummer

, Lui

, Lladó

, Lewczuk

, Li

, Martins

, Masters

, McAuliffe

, Mercken

, Moghekar

, Molinuevo

, Montine

, Nowatzke

, O’Brien

, Otto

, Paraskevas

, Parnetti

, Petersen

, Prvulovic

, de Reus

, Rissman

, Scarpini

, Stefani

, Soininen

, Schröder

, Shaw

, Skinningsrud

, Skrogstad

, Spreer

, Talib

, Teunissen

, Trojanowski

, Tumani

, Umek

, Van Broeck

, Vanderstichele

, Vecsei

, Verbeek

, Windisch

, Zhang

, Zetterberg

, Blennow

(2011) The Alzheimer’s Association external quality control program for cerebrospinal fluid biomarkers. Alzheimers Dement 7, 386–395.e6.

18.

Teunissen

, Tumani

, Engelborghs

, Mollenhauer

(2014) Biobanking of CSF: International standardization to optimize biomarker development. Clin Biochem 47, 288–292.

19.

Leitão

, Silva-Spínola

, Santana

, Olmedo

, Nadal

, Le Bastard

, Baldeiras

(2019) Clinical validation of the Lumipulse G cerebrospinal fluid assays for routine diagnosis of Alzheimer’s disease. Alzheimers Res Ther 11, 91.

20.

Verberk

IMW

, Misdorp

, Koelewijn

, Ball

, Blennow

, Dage

, Fandos

, Hansson

, Hirtz

, Janelidze

, Kang

, Kirmess

, Kindermans

, Lee

, Meyer

, Shan

, Shaw

, Waligorska

, West

, Zetterberg

, Edelmayer

, Teunissen

(2022) Characterization of pre-analytical sample handling effects on a panel of Alzheimer’s disease-related blood-based biomarkers: Results from the Standardization of Alzheimer’s Blood Biomarkers (SABB) working group. Alzheimers Dement 18, 1484–1497.

21.

Jack

Jr , Bennett

, Blennow

, Carrillo

, Feldman

, Frisoni

, Hampel

, Jagust

, Johnson

, Knopman

, Petersen

, Scheltens

, Sperling

, Dubois

(2016) A/T/N: An unbiased descriptive classification scheme for Alzheimer disease biomarkers. Neurology 87, 539–547.

22.

De Meyer

, Shapiro

, Vanderstichele

, Vanmechelen

, Engelborghs

, De Deyn

, Coart

, Hansson

, Minthon

, Zetterberg

, Blennow

, Shaw

, Trojanowski

; Alzheimer’s Disease Neuroimaging Initiative (2010) Diagnosis-independent Alzheimer disease biomarker signature in cognitively normal elderly people. Arch Neurol 67, 949–956.

23.

Fischl

, Salat

, Busa

, Albert

, Dieterich

, Haselgrove

, van der Kouwe

, Killiany

, Kennedy

, Klaveness

, Montillo

, Makris

, Rosen

, Dale

(2002) Whole brain segmentation: Automated labeling of neuroanatomical structures in the human brain. Neuron 33, 341–355.

24.

Tzourio-Mazoyer

, Landeau

, Papathanassiou

, Crivello

, Etard

, Delcroix

, Mazoyer

, Joliot

(2002) Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. Neuroimage 15, 273–289.

25.

Folstein

, Folstein

, McHugh

(1975) “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198.

26.

Peña-Casanova

, Gramunt-Fombuena

, Quiñones-Ubeda

, Sánchez-Benavides

, Aguilar

, Badenes

, Molinuevo

, Robles

, Barquero

, Payno

, Antúnez

, Martínez-Parra

, Frank-García

, Fernández

, Alfonso

, Sol

, Blesa

; NEURONORMA Study Team (2009) Spanish Multicenter Normative Studies (NEURONORMA Project): Norms for the Rey-Osterrieth complex figure (copy and memory), and free and cued selective reminding test. Arch Clin Neuropsychol 24, 371–393.

27.

Wechsler

(1945) A standardized memory scale for clinical use. J Psychol 19, 87–95.

28.

Alegret

, Valero

, Ortega

, Espinosa

, Sanabria

, Hernández

, Rodríguez

, Rosende-Roca

, Mauleón

, Vargas

, Martín

, Ruíz

, Tárraga

, Amariglio

, Rentz

, Boada

(2015) Validation of the Spanish Version of the Face Name Associative Memory Exam (S-FNAME) in Cognitively Normal Older Individuals. Arch Clin Neuropsychol 30, 712–720.

29.

Tombaugh

(2004) Trail Making Test A and B: Normative data stratified by age and education. Arch Clin Neuropsychol 19, 203–214.

30.

Papp

, Rentz

, Orlovsky

, Sperling

, Mormino

(2017) Optimizing the preclinical Alzheimer’s cognitive composite with semantic processing: The PACC5. Alzheimers Dement (N Y) 3, 668–677.

31.

Karikari

, Pascoal

, Ashton

, Janelidze

, Benedet

, Rodriguez

, Chamoun

, Savard

, Kang

, Therriault

, Schöll

, Massarweh

, Soucy

, Höglund

, Brinkmalm

, Mattsson

, Palmqvist

, Gauthier

, Stomrud

, Zetterberg

, Hansson

, Rosa-Neto

, Blennow

(2009) 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.

32.

Lobo

, Launer

, Fratiglioni

, Andersen

, Di Carlo

, Breteler

, Copeland

, Dartigues

, Jagger

, Martinez-Lage

, Soininen

, Hofman

(2000) Prevalence of dementia and major subtypes in Europe: A collaborative study of population-based cohorts. Neurologic Diseases in the Elderly Research Group. Neurology 54(11 Suppl 5), S4–9.

33.

Smirnov

, Ashton

, Blennow

, Zetterberg

, Simrén

, Lantero-Rodriguez

, Karikari

, Hiniker

, Rissman

, Salmon

, Galasko

(2022) Plasma biomarkers for Alzheimer’s disease in relation to neuropathology and cognitive change. Acta Neuropathol 143, 487–503.

34.

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.

35.

Tropea

, Waligorska

, Xie

, Nasrallah

, Cousins

KAQ

, Trojanowski

, Grossman

, Irwin

, Weintraub

, Lee

, Wolk

, Chen-Plotkin

, Shaw

(2023) Alzheimer’s Disease Neuroimaging Initiative. Plasma phosphorylated tau181 predicts cognitive and functional decline. Ann Clin Transl Neurol 10, 18–31.

36.

Ashton

, Janelidze

, Mattsson-Carlgren

, Binette

, Strandberg

, Brum

, Karikari

, González-Ortiz

, Di Molfetta

, Meda

, Jonaitis

, Koscik

, Cody

, Betthauser

, Li

, Vanmechelen

, Palmqvist

, Stomrud

, Bateman

, Zetterberg

, Johnson

, Blennow

, Hansson

(2022) Differential roles of Aβ42/40, p-tau231 and p-tau217 for Alzheimer’s trial selection and disease monitoring. Nat Med 28, 2555–2562.

37.

Tissot

, L Benedet

, Therriault

, Pascoal

, Lussier

, Saha-Chaudhuri

, Chamoun

, Savard

, Mathotaarachchi

, Bezgin

, Wang

, Fernandez Arias

, Rodriguez

, Snellman

, Ashton

, Karikari

, Blennow

, Zetterberg

, De Villers-Sidani

, Huot

, Gauthier

, Rosa-Neto

; Alzheimer’s Disease Neuroimaging Initiative (2021) Plasma pTau181 predicts cortical brain atrophy in aging and Alzheimer’s disease. Alzheimers Res Ther 13, 69.

38.

Bilgel

, Wong

, Moghekar

, Ferrucci

, Resnick

; Alzheimer’s Disease Neuroimaging Initiative (2022) Causal links among amyloid, tau, and neurodegeneration. Brain Commun 4, fcac193.

39.

Aguillon

, Langella

, Chen

, Sanchez

, Su

, Vila-Castelar

, Vasquez

, Zetterberg

, Hansson

, Dage

, Janelidze

, Chen

, Fox-Fuller

, Aduen

, Martinez

, Garcia

, Baena

, Guzman

, Johnson

, Sperling

, Blennow

, Reiman

, Lopera

, Quiroz

(2023) Plasma p-tau217 predicts in vivo brain pathology and cognition in autosomal dominant Alzheimer’s disease. Alzheimers Dement 19, 2585–2594.

40.

Chatterjee

, Pedrini

, Doecke

, Thota

, Villemagne

, Doré

, Singh

, Wang

, Rainey-Smith

, Fowler

, Taddei

, Sohrabi

, Molloy

, Ames

, Maruff

, Rowe

, Masters

, Martins

; AIBL Research Group (2023) Plasma Aβ42/40 ratio, p-tau181, GFAP, and NfL across the Alzheimer’s disease continuum: A cross-sectional and longitudinal study in the AIBL cohort. Alzheimers Dement 19, 1117–1134.