sTREM2 in discordant CSF Aβ42 and p‐tau181

Danni Li; William G Mantyh; Lu Men; Ishika Jain; Matthew Glittenberg; Binchong An; Lin Zhang; Ling Li; for the Alzheimer's Disease Neuroimaging Initiative

doi:10.1002/dad2.70072

. 2025 Jan 16;17(1):e70072. doi: 10.1002/dad2.70072

sTREM2 in discordant CSF Aβ₄₂ and p‐tau181

Danni Li ^1,^✉, William G Mantyh ², Lu Men ¹, Ishika Jain ¹, Matthew Glittenberg ³, Binchong An ¹, Lin Zhang ⁴, Ling Li ⁵; for the Alzheimer's Disease Neuroimaging Initiative

PMCID: PMC11736621 PMID: 39822286

Abstract

INTRODUCTION

Little is known about the factors underpinning discordant cerebrospinal fluid (CSF) amyloid beta (Aβ)₄₂ versus p‐tau181/Aβ₄₂ or CSF Aβ₄₂ versus Aβ positron emission tomography (PET).

METHODS

We stratified 570 non‐demented Alzheimer's Disease Neuroimaging Initiative (ADNI) participants by Aβ PET and further by CSF Aβ₄₂ or p‐tau181/Aβ₄₂. We used analysis of covariance testing adjusting for covariates, followed by Tukey post hoc pairwise comparisons, to compare CSF soluble triggering receptor expressed on myeloid cells‐2 (sTREM2) across four participant groups: CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42}, CSF− _Aβ42 with CSF+ _{p‐tau/Aβ42}, and concordant CSF_Aβ42/CSF_{p‐tau/Aβ42}. We also compared sTREM2 across discordant and concordant CSF_Aβ42/PET.

RESULTS

Regardless of Aβ PET status, CSF+_Aβ42 with CSF−_{p‐tau/Aβ42} had lower sTREM2 than CSF−_Aβ42 with CSF+_{p‐tau/Aβ42}. CSF sTREM2 was similarly also associated with discordant CSF Aβ42 /PET.

DISCUSSION

Our study suggests the potential roles of sTREM2 in discordant CSF Aβ₄₂ and p‐tau181/Aβ₄₂ and discordant CSF_Aβ42/PET. Low‐ and high‐CSF sTREM2 may affect the accuracy of p‐tau181/Aβ₄₂ during the clinical work‐up of AD.

Highlights

17% of non‐demented older adults had discordant CSF Aβ₄₂ versus p‐tau181/Aβ₄₂.
sTREM2 differed between discordant cases of CSF Aβ₄₂ versus p‐tau181/Aβ₄₂.
20% of non‐demented older adults had discordant CSF Aβ₄₂ versus Aβ PET.
sTREM2 also differed between discordant cases of CSF Aβ₄₂ versus Aβ PET.
p‐tau181/Aβ₄₂ may miss 6.7% of PET+ non‐demented older adults with low sTREM2.

Keywords: Alzheimer's disease (AD), amyloid‐β PET, CSF_Aβ42 , CSF_{p‐tau/Aβ42} , CSF sTREM2

1. BACKGROUND

Alzheimer's disease (AD) is now a biologically defined entity that follows a stereotypical “pathophysiological cascade” starting with abnormal amyloid beta (Aβ) followed by abnormal phosphorylated tau (p‐tau), neurodegeneration, and clinical symptoms. ¹ A recent longitudinal study demonstrated that cerebrospinal fluid (CSF) Aβ₄₂ and Aβ_42/40 ratio diverged 18 and 14 years before symptom onset, respectively, between those who developed AD later and those who did not, compared to 11 years of CSF p‐tau181, confirming that changes in CSF Aβ₄₂ and Aβ_42/40 ratio precede p‐tau181. ² Despite the temporal differences, Aβ₄₂ and p‐tau181 are frequently used separately for confirming underlying AD pathology. Specifically, FDA‐approved tests employ ratios of CSF Aβ₄₂ and p‐tau181 (ie, CSF Aβ₄₂/₄₀ and p‐tau181/Aβ₄₂), which were shown to be superior to the individual biomarkers when compared to Aβ positron emission tomography (PET). ³ CSF Aβ₄₂/₄₀ and p‐tau181/Aβ₄₂ were equally predictive of amyloid‐β PET status in participants from the Mayo Clinic Study of Aging and their associated Alzheimer's Disease Research Center. ⁴ However, it is not infrequent for patients to have discordant results with these CSF ratios or individual biomarkers (CSF Aβ_42, Aβ_42/40, p‐tau181, p‐tau181/Aβ₄₂). ⁵ , ⁶ Furthermore, some patients exhibit discordant results between CSF Aβ₄₂ and Aβ PET. ⁷

In this study, we investigated CSF soluble triggering receptor expressed on myeloid cells 2 (sTREM2) between discordant cases of CSF Aβ₄₂ versus p‐tau181/Aβ₄₂ and of CSF Aβ₄₂ versus Aβ PET. CSF sTREM2 is a surrogate marker for microglial activation that plays a crucial role in AD development. While the roles of TREM2 and sTREM2 in the pathophysiology of AD are complex and still a matter of active research, an emerging body of knowledge shows that increased sTREM2 is associated with slowed rates of atrophy and clinical progression. ⁸ , ⁹ , ¹⁰ , ¹¹ TREM2 is crucial for microglia to compact Aβ, leading to plaques that, in turn, limit the spread of Aβ, preserve the neurites of nearby neurons, and ultimately may play a neuroprotective role. ¹² sTREM2 results from the proteolytic shedding of TREM2 and is known to activate microglia, bind oligomeric Aβ₄₂, and protect against Aβ pathology in mice. ¹³ Recent studies suggest the roles of high sTREM2 in attenuating the decrease of CSF Aβ₄₂ while simultaneously increasing p‐tau181 in patients with abnormal Aβ, ⁸ , ¹⁴ pointing to the relevance of sTREM2 for the interpretation of CSF Aβ₄₂ and p‐tau181 in the clinical setting. The current work thus examines CSF sTREM2 in discordant CSF Aβ₄₂ and p‐tau181/Aβ₄₂ and discordant CSF Aβ₄₂ and Aβ PET in non‐demented participants in ADNI. We use Aβ₄₂ instead of Aβ₄₂/₄₀ because ADNI has limited Aβ₄₀ data to calculate the ratio. Although Aβ₄₂ is not used clinically for diagnosing AD, Aβ₄₂ is superior to Aβ₄₂/₄₀ for detecting earlier changes, ² and CSF Aβ₄₂ alone has been widely used in research studies to identify AD. We use p‐tau181/Aβ₄₂ because it is clinically used for diagnosis of AD.

2. METHODS

2.1. Study participants

Data used to prepare this article were obtained from the ADNI database (adni.loni.usc.edu). The ADNI was launched in 2003 as a public–private partnership led by Principal Investigator Michael W. Weiner, MD. The primary goal of ADNI has been to test whether serial magnetic resonance imaging (MRI), PET, fluid biomarkers, and clinical and neuropsychological assessment could be combined to measure the progression of mild cognitive impairment (MCI) and early AD.

We included ADNI participants without dementia who were cognitively normal (CN) or had MCI. Participants were included if they also had available (1) CSF Aβ₄₂ and CSF p‐tau181, (2) CSF sTREM2, (3) Aβ PET, (4) demographic/medical information (age, sex, and body mass index [BMI]), and (5) the ε4 allele of the apolipoprotein E gene (APOE4) carrier status. CN participants had Mini‐Mental State Examination (MMSE) scores between 24 and 30 and a global Clinical Dementia Rating (CDR) of 0. ¹⁵ MCI participants had MMSE scores between 24 and 30, a global CDR score of 0.5, a CDR memory score of 0.5 or greater, and objective memory impairment on the Wechsler Memory Scale – Logical Memory II test. ¹⁵ Participants were excluded if they had (1) a diagnosis of dementia documented at their baseline visit and (2) a time difference exceeding 180 days between the dates documented for Aβ PET and CSF biomarker (Aβ₄₂, sTREM2) examinations. Applying these criteria resulted in a final study population of 570 participants. Table S1 lists the data files used for biomarker variables.

Ethics approval was obtained by the ADNI investigators from the local ethics committees of all involved sites, and informed consent was obtained from all participants or legal guardian(s)/legally authorized representatives. All methods were carried out according to relevant guidelines and regulations. Per the Institutional Review Board of the University of Minnesota, this study was non‐human subject research.

2.2. AD biomarkers

CSF Aβ₄₂ and p‐tau181 were assessed by a fully automated Elecsys Cobas e 601 instrument by the ADNI Biomarker Core and CSF sTREM2 by Meso Scale Discovery ELISA by Christian Haass's lab. ⁹ , ¹⁶ , ¹⁷ ADNI performed ¹⁸F‐florbetapir PET acquisition and analysis according to previous protocols. ¹⁸ , ¹⁹ Specifically, ADNI used a native‐space structural MRI scan that was closest in time to each Aβ PET scan and first segmented and parcellated the MRI scan with Freesurfer (FS) version 7.1.1 and then coregistered to the Aβ PET imaging with statistical parametric mapping (SPM). The Aβ PET measurement reflected the tracer uptake in a cortical summary standardized uptake value ratio (SUVR) (ie, composed of frontal, anterior/posterior cingulate, lateral parietal, and lateral temporal regions, defined by the Desikan–Killiany atlas) and captured the overall amyloid burden in the brain. ADNI normalized PET intensity using the whole cerebellum. This study used the ADNI‐recommended 1.11 cutoff to determine amyloid positivity based on the upper limit of cortical uptake in whole cerebellum‐normalized SUVRs in young control samples. ²⁰

We defined CSF Aβ₄₂ positivity as <976.6 pg/mL, which was optimized for Aβ PET concordance, ²¹ in the primary analysis. To ensure the robustness of the study results, we used two different Aβ₄₂ cut points (<880 pg/mL and <1100 pg/mL), which were not optimized for Aβ PET concordance in ADNI. Specifically, an Aβ₄₂ cut point <1100 pg/mL was optimized for the BioFINDER cohort; a cut point <880 pg/mL was based on 1100 pg/mL and an adjustment factor of 0.8, which accounted for pre‐analytical differences from the BioFINDER to the ADNI cohort. ²¹

The CSF p‐tau181/Aβ₄₂ or Aβ_42/40 ratio is superior to CSF Aβ₄₂ alone for identifying patients with AD, primarily due to the robustness of the ratios against the impact of pre‐analytical factors and differences in the processing of amyloid precursor protein. ²² The CSF p‐tau181/Aβ₄₂ ratio was calculated based on the Roche Elecsys CSF p‐tau181 and Aβ₄₂ tests reported in their original units of ng/mL and pg/mL, respectively. A CSF p‐tau181/Aβ₄₂ ratio >0.023 is FDA‐approved to confirm underlying AD pathology (ie, consistent with a positive Aβ PET scan result). ⁴ Because the Roche Elescys CSF Aβ₄₂ test has an upper reportable limit of 1700 pg/mL, we assigned a value of 1701 pg/mL to the Elecsys Aβ₄₂ values >1700 pg/mL for the CSF p‐tau181/Aβ₄₂ calculations. ⁴

RESEARCH CONTEXT

Systematic review: We reviewed the literature using Google Scholar and found that despite the many existing studies of CSF sTREM2, none examined whether CSF sTREM2 underpins discordant CSF Aβ₄₂ versus p‐tau181/Aβ₄₂ or CSF Aβ₄₂ versus Aβ PET.
Interpretation: Our findings demonstrated a substantial minority of non‐demented older adults had discrepant CSF results according to CSF Aβ₄₂ versus p‐tau181/Aβ₄₂ ratio or CSF Aβ₄₂ versus Aβ PET, which were associated with opposite CSF sTREM2 levels. Our results are consistent with prior work examining CSF sTREM2, Aβ₄₂, and p‐tau181 and suggest low and high sTREM2 may affect the accuracy of using the p‐tau181/Aβ₄₂ ratio for clinical diagnosis of AD.
Future directions: We want to study the possibility of adding CSF sTREM2 to improve the diagnostic accuracy of AD biomarkers and understand the biological factors contributing to differences in CSF sTREM2.

2.3. MRI scans

Prior research suggested abnormal CSF flow dynamics as an explanation for false‐positive CSF Aβ₄₂ but normal p‐tau181/Aβ₄₂ ratio ²³ , ²⁴ or normal Aβ PET. ²⁵ A board‐certified neurologist with a subspecialty training in behavioral/cognitive neurology (WGM) visually evaluated 41 MRI scans of PET− participants with CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42} and 18 MRI scans of PET+ participants with CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42} for findings of disproportionate enlargement of subarachnoid space (DESH). ²⁶

2.4. Statistical analysis

ANOVA was used to compare continuous variables (eg, age, Aβ PET SUVR, CSF Aβ₄₂, CSF p‐tau181/Aβ₄₂, and CSF sTREM2) and reported means and standard deviations and chi‐squared tests to compare categorical variables (eg, cognitive status, sex, and APOE4 presence) across biomarker categories and reported counts and percentages in Table 1. An analysis of covariance (ANCOVA) test with adjustment for age, sex, APOE4 status, and cognitive status followed by Tukey post hoc pairwise comparisons was used to determine whether CSF sTREM2 between any two groups was significantly different (Tukey adjusted p < .05).

TABLE 1.

Demographic and data across biomarker categories based on CSF Aβ₄₂ and PET (left) and CSF p‐tau181/Aβ₄₂ and PET (right).

		CSF Aβ42					CSF p‐tau181/Aβ42
		PET− (301)		PET+ (269)			PET− (301)		PET+ (269)
	Total	CSF−/PET−	CSF+/PET−	CSF−/PET+	CSF+/PET+	p value	CSF−/PET−	CSF+/PET−	CSF−/PET+	CSF+/PET+	p value
N	570	246	55	57	212		277	24	47	222
Age (year)	71.9 (7.0)	70.8 (7.0)	70.0 (7.2)	72.8 (7.1)	73.3 (6.6)	<.001	70.3 (7.0)	74.2 (6.9)	72.2 (6.4)	73.4 (6.8)	<.001
BMI (kg/m²)	27.6 (4.8)	28.3 (4.8)	27.3 (4.9)	27.4 (5.2)	26.8 (4.5)	.0090	28.3 (4.9)	26.4 (4.4)	28.2 (5.4)	26.7 (4.5)	.0011
Sex
Female (n)	266 (47%)	119 (48%)	17 (31%)	37 (65%)	93 (44%)	.0028	127 (46%)	9 (38%)	29 (62%)	101 (45%)	.1527
Male (n)	304 (53%)	127 (52%)	38 (69%)	20 (35%)	119 (56%)		150 (54%)	15 (62%)	18 (38%)	121 (55%)
Cognitive status
CN (n)	205 (36%)	115 (47%)	22 (40%)	21 (37%)	47 (22%)	<.001	129 (47%)	8 (33%)	22 (47%)	46 (21%)	<.001
MCI (n)	365 (64%)	131 (53%)	33 (60%)	36 (63%)	165 (78%)		148 (53%)	16 (67%)	25 (53%)	176 (79%)
APOE4 carrier status
No (n)	331 (58%)	197 (80%)	34 (62%)	37 (65%)	63 (30%)	<.001	218 (79%)	13 (54%)	33 (70%)	67 (30%)	<.001
Yes (n)	239 (42%)	49 (20%)	21 (38%)	20 (35%)	149 (70%)		59 (21%)	11 (46%)	14 (30%)	155 (70%)
CSF Aβ42 (pg/mL)	1104.6 (447.9)	1513.4 (236.6)	780.0 (157.7)	1277.9 (249.6)	667.9 (176.0)	<.001	1421.2 (329.9)	897.4 (369.9)	1169.4 (375.5)	718.4 (236.5)	<.001
Amyloid PET (SUVR)	1.2 (0.2)	1.0 (0.1)	1.0 (0.1)	1.3 (0.2)	1.4 (0.2)	<.001	1.0 (0.1)	1.1 (0.1)	1.2 (0.1)	1.4 (0.2)	<.001
CSF p‐tau (ng/mL)	25.0 (13.4)	19.5 (6.7)	15.4 (7.6)	32.0 (15.9)	32.1 (15.1)	<.001	17.8 (5.9)	29.5 (10.1)	18.9 (6.1)	34.9 (15.1)	<.001
CSF p‐tau/Aβ42¹	29.3 (24.9)	13.0 (4.6)	21.6 (14.2)	26.5 (15.4)	51.0 (27.2)	<.001	12.7 (3.4)	35.9 (13.6)	16.7 (4.1)	52.0 (25.8)	<.001
CSF sTREM2 (pg/mL)	3904.5 (2123.6)	3907.7 (2066.8)	2694.6 (1417.6)	5075.4 (2222.2)	3899.7 (2150.5)	<.001	3552.7 (1830.7)	5225.3 (3192.5)	4019.3 (2105.1)	4176.3 (2240.7)	<.001

Open in a new tab

Note: For continuous variables, means, standard deviations, and p values from ANOVAs are reported; for categorical variables, counts, percentages, and p from chi‐squared tests are reported. CSF p‐tau/Aβ42 ratio was calculated using the original unit (ie, [ng/mL]/[pg/mL]), and then the results were scaled by 1000 to avoid preceding zeros. Therefore, the FDA‐approved cut point of >0.023 using the original unit (ie, [ng/mL]/[pg/mL]) should be scaled accordingly by 1000 to be >23.

Abbreviation: sTREM2, soluble triggering receptor expressed on myeloid cells 2; SUVR, standardized uptake value ratio.

3. RESULTS

3.1. Categorization of study participants based on CSF Aβ42 or p‐tau181/Aβ42 and Aβ PET

We first divided the 570 non‐demented ADNI participants into Aβ PET− (n = 301) and PET+ (n = 269). We used the ADNI‐recommended >1.11 cutoff to determine Aβ PET positivity. ¹⁸ , ²⁰ We then divided each PET group into CSF− and CSF+ based on Aβ₄₂ or p‐tau181/Aβ₄₂. For the CSF Aβ₄₂ cut point, we used <976.6 pg/mL, determined to achieve optimal concordance with Aβ PET in ADNI. ²¹ For CSF p‐tau181/Aβ₄₂, we used >0.023, which is approved by the FDA for identifying AD. Using the combination of CSF and PET biomarkers, we ended up with four groups (CSF−/PET−, CSF+/PET−, CSF−/PET+, CSF+/PET+) for each of the Aβ₄₂ and p‐tau181/Aβ₄₂ CSF categorizations. Table 1 lists the demographics of four biomarker categories based on CSF Aβ₄₂ and PET (left) and CSF p‐tau181/Aβ₄₂ and PET (right). Of 301 PET− participants, 246 and 55 were classified based on Aβ₄₂ as CSF− and CSF+, respectively; 277 and 24 were classified as CSF− and CSF+ based on p‐tau181/Aβ₄₂, respectively. Of 269 PET+ participants, 57 and 212 were classified based on Aβ₄₂ as CSF− and CSF+, respectively; 47 and 222 were classified as CSF− and CSF+ based on p‐tau181/Aβ_42, respectively.

Of the 570 participants, 19.6% (112/570) had CSF/PET discordant groups (ie, CSF+/PET− and CSF‐/PET+) based on CSF Aβ₄₂, higher than 12.4% (or 71/570) of participants with CSF/PET discordant groups based on CSF p‐tau181/Aβ₄₂. The CSF/PET concordant groups (ie, CSF−/PET− and CSF+/PET+) based on CSF Aβ₄₂ were similar to the CSF/PET concordant groups based on CSF p‐tau181/Aβ₄₂ in that both age and percentage of participants with MCI gradually increased in parallel to CSF and PET positivity, with CSF−/PET− demonstrating lower and CSF+/PET+ demonstrating higher age and percentage of participants with MCI. The CSF/PET discordant groups (ie, CSF+/PET− and CSF−/PET+) based on CSF Aβ₄₂ were similar to the CSF/PET discordant groups based on CSF p‐tau181/Aβ₄₂ in that there were more male in the CSF+/PET− groups than in the CSF−/PET+ groups. However, age between the discordant CSF/PET groups had reverse directions: based on CSF Aβ₄₂, CSF+/PET− participants were younger than CSF−/PET+ participants (70.0 [7.2] years vs 72.8 [7.1] years), whereas, based on p‐tau181/Aβ₄₂, CSF+/PET− participants were older than CSF−/PET+ participants (74.2 [6.9] years vs 72.2 [6.4] years).

3.2. sTREM2 in participants with discordant CSF Aβ₄₂ and p‐tau181/Aβ₄₂

Table 2 summarizes the participants who had different CSF classifications depending on Aβ₄₂ versus p‐tau181/Aβ₄₂, along with demographics, CSF p‐tau181, and CSF sTREM2 (Table 2). Among the 301 PET− participants, 41 had CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42}. These participants had low CSF p‐tau181 (mean [SD]: 12.1[3.1] pg/mL), despite low Aβ₄₂. Ten PET− participants had CSF− _Aβ42 with CSF+ _{p‐tau/Aβ42}. These participants had high CSF p‐tau181 (35.8 [8.6] pg/mL), despite normal Aβ₄₂ (Figure 1A,B,C). The 41 CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42} participants had lower sTREM2 than 10 CSF− _Aβ42 with CSF+ _{p‐tau/Aβ42} participants (2245.1 [1086.9] vs 6925.1 [4165.6] pg/mL, Tukey adjusted p value < .0001, by ANCOVA test adjusting for age, sex, APOE4 status, and cognitive status followed by Tukey post hoc pairwise group comparisons) (Table 2 and Figure 1D). CSF p‐tau181/Aβ₄₂ classified 41 more PET− participants correctly compared to Aβ₄₂ but classified 10 more PET− participants incorrectly compared to Aβ₄₂ (Figure 1A).

TABLE 2.

Demographics and data of participants who had different CSF classifications depending on Aβ₄₂ versus p‐tau181/Aβ₄₂ (CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42} and CSF− _Aβ42 with CSF+ _{p‐tau/Aβ42}).

	PET− (51/301)		PET+ (46/269)
	CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42}	CSF− _Aβ42 with CSF+ _{p‐tau/Aβ42}	CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42}	CSF− _Aβ42 with CSF+ _{p‐tau/Aβ42}	p value
N	41	10	18	28
Age (year)	68.9 (7.7)	75.9 (9.5)	76.3 (6.0)	75.9 (7.5)	<.001
BMI (kg/m2)	27.7 (4.8)	26.7 (2.6)	28.7 (4.9)	27.0 (4.6)	.7897
Sex
Female (n)	12 (29%)	4 (40%)	7 (39%)	15 (54%)	.1594
Male (n)	29 (71%)	6 (60%)	11 (61%)	13 (46%)
Cognitive status
CN (n)	16 (39%)	2 (20%)	10 (56%)	9 (32%)	.3435
MCI (n)	25 (61%)	8 (80%)	8 (44%)	19 (68%)
APOE4 carrier Status
No (n)	27 (66%)	6 (60%)	12 (67%)	16 (57%)	.7759
Yes (n)	14 (34%)	4 (40%)	6 (33%)	12 (43%)
CSF Aβ42 (pg/mL)	823.8 (123.1)	1241.2 (277.3)	783.2 (132.9)	1142.1 (153.6)	<.001
Amyloid PET (SUVR)	1.0 (0.1)	1.0 (0.1)	1.2 (0.1)	1.4 (0.2)	<.001
CSF p‐tau (ng/mL)	12.1 (3.1)	35.8 (8.6)	14.1 (3.6)	42.5 (16.5)	<.001
CSF p‐tau/Aβ42¹	15.0 (4.1)	29.0 (5.0)	18.2 (4.1)	37.5 (15.0)	<.001
CSF sTREM2 (pg/mL)	2245.1 (1086.9)	6925.1 (4165.6)	3465.5 (2428.8)	5813.2 (2372.3)	<.001

Open in a new tab

Abbreviation: sTREM2, soluble triggering receptor expressed on myeloid cells 2; SUVR, standardized uptake value ratio.

CSF p‐tau181 and sTREM2 within PET− participants who had different CSF classifications depending on Aβ₄₂ versus p‐tau181/Aβ₄₂. (A) Illustration of PET− participants who were CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42} (green) and CSF− _Aβ42 with CSF+ _{p‐tau/Aβ42} (red). (B) CSF p‐tau181 concentrations of CSF− and CSF+ based on Aβ₄₂; an ANOVA test was used to determine whether p‐tau181 differed between CSF− and CSF+. (C) CSF p‐tau181 concentrations of CSF− and CSF+ based on p‐tau181/Aβ₄₂; an ANOVA test was used to determine whether p‐tau181 differed between CSF− and CSF+. (D) CSF sTREM2 concentrations of participants who were CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42} (green), CSF− _Aβ42 with CSF+ _{p‐tau/Aβ42} (red), concordant CSF− by Aβ₄₂ and p‐tau/Aβ₄₂ (light gray), and concordant CSF+ by Aβ₄₂ and p‐tau/Aβ₄₂ (dark gray); an analysis of covariance test with adjustment for age, sex, *APOE4* status, and cognitive status followed by Tukey post hoc pairwise comparisons was used to determine whether CSF sTREM2 between any two groups was significantly different (Tukey adjusted p value < .05). *: .01 < p or Tukey adjusted p value < .05; **: .001 < p or Tukey adjusted p value < .05; ***: 0.0001 < p or Tukey adjusted p value < .001; ****: p or Tukey adjusted p value < .0001. sTREM2, soluble triggering receptor expressed on myeloid cells‐2.

Among the 269 PET+ participants, 18 had CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42}. These participants had low p‐tau181 (14.1[3.6] pg/mL), despite decreased Aβ₄₂. Twenty‐eight PET+ participants had CSF− _Aβ42 with CSF+ _{p‐tau/Aβ42}. These participants had high CSF p‐tau181 (42.5 [16.5] pg/mL), despite normal Aβ₄₂ (Figure 2A,B,C). The 18 CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42} participants had lower sTREM2 than the 28 CSF− _Aβ42 with CSF+ _{p‐tau/Aβ42} participants (3465.5[2428.8] vs 5813.2 [2372.3] pg/mL, Tukey adjusted p value < .001) (Table 2 and Figure 2D). P‐tau181/Aβ₄₂ classified 28 more PET+ participants correctly compared to Aβ₄₂, but it classified 18 more PET+ participants incorrectly compared to Aβ₄₂ (Figure 2A).

Although a decrease in CSF Aβ₄₂ is the earliest pathology of AD, and CSF Aβ₄₂ alone has been widely used in research studies to identify AD, there is no consensus on which CSF Aβ₄₂ cut points should be used to diagnose AD. To ensure our results were robust for different CSF Aβ₄₂ cut points, we used two more CSF Aβ₄₂ cut points: 880 and 1100 pg/mL that were not optimized for Aβ PET concordance in ADNI. Figure S1 demonstrated that the sTREM2 results were similar when different CSF Aβ₄₂ cut points were used, in that CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42} participants had lower sTREM2 than CSF− _Aβ42 to CSF+ _{p‐tau/Aβ42,} regardless of amyloid PET status (ie, PET− or PET+).

3.3. sTREM2 in participants with discordant CSF Aβ₄₂ and Aβ PET

In addition to discordant CSF Aβ₄₂ and p‐tau181/Aβ₄₂, we investigated CSF sTREM2 levels in discordant CSF Aβ₄₂ and Aβ PET (left side of Table 1). Of the 570 participants, 112 had discordant CSF Aβ₄₂ and Aβ PET: 55 were CSF+/PET− and 57 CSF−/PET+. The 55 CSF+/PET− participants had lower sTREM2 levels than the 57 CSF−/PET+ participants (2694.6 [1417.6] vs 5075.4 [2222.2] pg/mL, Tukey adjusted p value < .001) (Figure 3). Figure S2 demonstrates that the sTREM2 results were similar when different CSF Aβ₄₂ cut points (< 880 pg/mL and < 1100 pg/mL) were used, in that CSF+/PET− participants had lower sTREM2 than CSF−/PET+ participants.

CSF sTREM2 levels across biomarker categories based on CSF Aβ₄₂ (< 976.6 pg/mL) and Aβ PET. An analysis of covariance test with adjustment for age, sex, *APOE4* status, and cognitive status followed by Tukey post hoc pairwise comparisons was used to determine whether CSF sTREM2 between any two groups was significantly different (Tukey adjusted p value < .05). *Tukey adjusted p value < .05; **: .001 ≤ Tukey's adjusted p value < .01; ***: Tukey adjusted p value < .001. ns: not significant; sTREM2, soluble triggering receptor expressed on myeloid cells 2.

4. DISCUSSION

We found that there were more participants with CSF/PET discordant based on CSF Aβ₄₂ and PET than based on CSF p‐tau181/Aβ₄₂ and PET (19.6% [or 112/570] vs 12.4% [or 71/570]), consistent with previous findings that p‐tau181/Aβ₄₂ was superior to Aβ₄₂ when compared to Aβ PET. ³ The study's primary finding is that a substantial minority (approximately 17%) of non‐demented older adults had discrepant CSF results according to Aβ₄₂ versus the p‐tau181/Aβ₄₂ ratio, which was associated with opposite CSF sTREM2 levels. Participants who were classified as positive by Aβ₄₂ but negative by p‐tau181/Aβ₄₂ ratio (ie, CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42}) had lower sTREM2 than those participants who were classified as negative by Aβ₄₂ but positive by p‐tau181/Aβ₄₂ (ie, CSF− _Aβ42 with CSF+ _{p‐tau/Aβ42}), regardless of Aβ PET status (ie, PET− or PET+). Similarly, approximately 20% of non‐demented older adults had discordant CSF Aβ₄₂ and Aβ PET results, which was associated with opposite CSF sTREM2 levels, in that CSF+/PET− participants had lower sTREM2 levels than CSF−/PET+ participants. Of note, CSF sTREM2 is significantly associated with Aβ plaque‐related increase in CSF p‐tau181. ¹⁴ , ²⁷ In addition, a longitudinal study from the Dominantly Inherited Alzheimer Network (DIAN) cohort found that in early pre‐symptomatic stages of AD, an augmented annual rate of increase in sTREM2 was associated with a diminished annual rate decrease in CSF Aβ₄₂, consistent with a theory that microglia clustering around the smallest Aβ seeds at a very early stage of AD reduces the growth and spread of Aβ. ⁸ For participants with low sTREM2 and decreased CSF Aβ₄₂ with normal CSF p‐tau181/Aβ₄₂, insufficient sTREM2 may be implicated in abnormal Aβ₄₂ without concomitant increase in p‐tau181. For those with high sTREM2 with normal CSF Aβ₄₂ but abnormal CSF p‐tau181/Aβ₄₂, high sTREM2 may increase Aβ‐related p‐tau181 while attenuating the decrease of CSF Aβ₄₂. Our results have implications for the clinical diagnosis of AD in that the CSF p‐tau181/Aβ₄₂ ratio may miss the 6.7% (or 18/269) of PET+ non‐demented older adults with decreased Aβ₄₂, low p‐tau181, and low sTREM2 (ie, CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42}). In PET− participants, CSF p‐tau181/Aβ₄₂ ratio may misclassify the 3.3% (or 10/301) of PET− non‐demented older adults with normal Aβ₄₂, high p‐tau181, and high sTREM2 (ie, CSF− _Aβ42 with CSF+ _{p‐tau/Aβ42}).

A low CSF Aβ₄₂ but normal p‐tau181/Aβ42 (ie, CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42}) may be due to abnormal CSF dynamics such as normal pressure hydrocephalus. ²³ , ²⁴ We examined the MRI scans of CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42} participants to rule out abnormal CSF flow dynamics. Four of the 41 CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42} PET− participants and three of the 18 CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42} PET+ participants had one or more radiographic features of DESH (high and tight cerebral convexity, enlarged sylvian fissures, or ventriculomegaly disproportionate to cerebral atrophy). We conducted Fisher exact tests to compare the prevalence of DESH in our study to that of Graff‐Radford et al., who reported that 6.6% (or 45/684) of healthy participants had evidence of DESH per an automated MRI assessment. ²⁸ The p values exceeded .05, suggesting that the prevalence of DESH in our study of CSF+ _Aβ42 with CSF− _{p‐tau/Aβ42} PET− and PET+ participants were not statistically significant from that of Graff‐Radford et al.

This study's strength includes analyzing a comprehensive ADNI dataset to elucidate the role of CSF sTREM2 behind discordant CSF Aβ₄₂ and p‐tau181/Aβ₄₂ results and discordant CSF Aβ₄₂ and Aβ PET results. Due to limited CSF Aβ₄₀ data availability in ADNI, we were unable to assess whether there were any discrepant participants by CSF Aβ_42/40 and p‐tau181/Aβ₄₂ and whether CSF sTREM2 differed between discordant CSF Aβ_42/40 and p‐tau181/Aβ₄₂ results. Also, we do not know the underlying factors that influence CSF sTREM2 levels, of which there are many, such as multiple sclerosis, ²⁹ genetic variants (eg, rs7232, a missense variant in the MS4A6A gene of chromosome 11), and biological processes for regulating viruses and immune response. ³⁰ We cannot rule out non‐AD conditions that can increase p‐tau181, such as vascular dementia. ³¹ Another limitation is that ADNI's study population differs from that of community studies. Therefore, additional studies with more representation of diverse cohorts are needed to generalize and expand upon our findings.

5. CONCLUSIONS

This study demonstrated opposite CSF sTREM2 levels in discordant CSF Aβ₄₂ and p‐tau181/Aβ₄₂ and discordant CSF Aβ₄₂ and Aβ PET. Our results are consistent with prior work examining CSF sTREM2, Aβ₄₂, and p‐tau181 and suggest low and high sTREM2 levels may affect the accuracy of the p‐tau181/Aβ₄₂ ratio in the clinical diagnosis of AD.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest. Author disclosures are available in the Supporting Information.

CONSENT STATEMENT

Consent was not necessary for the study.

Supporting information

Supporting Information

DAD2-17-e70072-s002.docx^{(515.9KB, docx)}

Supporting Information

DAD2-17-e70072-s001.pdf^{(1.2MB, pdf)}

ACKNOWLEDGMENTS

Data collection and sharing for this project was funded by the Alzheimer's Disease Neuroimaging Initiative (ADNI) (National Institutes of Health [NIH] Grant U01 AG024904) and DOD ADNI (Department of Defense award number W81XWH‐12‐2‐0012). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: AbbVie, Alzheimer's Association; Alzheimer's Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol‐Myers Squibb Company; CereSpir, Inc.; Cogstate; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; EuroImmun; F. Hoffmann‐La Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.; Johnson & Johnson Pharmaceutical Research & Development LLC; Lumosity; Lundbeck; Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack Technologies; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Takeda Pharmaceutical Company; and Transition Therapeutics. The Canadian Institutes of Health Research provides funds to support ADNI clinical sites in Canada. Private‐sector contributions are facilitated by the Foundation for the NIH (www.fnih.org). The grantee organization is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer's Therapeutic Research Institute at the University of Southern California. The laboratory disseminates ADNI data for Neuro Imaging at the University of Southern California. This study is supported by the NIH (R01AG059654, RF1AG079100‐01A1, R01AG081426, R01AG080806).

Li D, Mantyh WG, Men L, et al. sTREM2 in discordant CSF Aβ₄₂ and p‐tau181. Alzheimer's Dement. 2025;17:e70072. 10.1002/dad2.70072

Data used to prepare this article were obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and provided data but did not participate in the analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp‐content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf.

DATA AVAILABILITY STATEMENT

All raw data used in this study are freely available from https://ida.loni.usc.edu.

REFERENCES

1. Jack CR, Jr ., Bennett DA, Blennow K, et al. NIA‐AA Research Framework: toward a biological definition of Alzheimer's disease. Alzheimers Dement. 2018;14(4):535‐562. doi: 10.1016/j.jalz.2018.02.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
2. Jia J, Ning Y, Chen M, et al. Biomarker changes during 20 years preceding Alzheimer's disease. N Eng J Med. 2024;390:712‐722. [DOI] [PubMed] [Google Scholar]
3. Leitão MJ, Silva‐Spínola A, Santana I, et al. Clinical validation of the Lumipulse G cerebrospinal fluid assays for routine diagnosis of Alzheimer's disease. Alzheimer's Research & Therapy. 2019;11:1‐12. [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Campbell MR, Ashrafzadeh‐Kian S, Petersen RC, et al. P‐tau/aβ42 and aβ42/40 ratios in csf are equally predictive of amyloid pet status. Alzheimers Dement. 2021;13(1):e12190. [DOI] [PMC free article] [PubMed] [Google Scholar]
5. De Wilde A, Reimand J, Teunissen CE, et al. Discordant amyloid‐β PET and CSF biomarkers and its clinical consequences. Alzheimers Res Ther. 2019;11(1):1‐13. [DOI] [PMC free article] [PubMed] [Google Scholar]
6. Vos SJ, Gordon BA, Su Y, et al. NIA‐AA staging of preclinical Alzheimer disease: discordance and concordance of CSF and imaging biomarkers. Neurobiol Aging. 2016;44:1‐8. [DOI] [PMC free article] [PubMed] [Google Scholar]
7. Sala A, Nordberg A, Rodriguez‐Vieitez E; Alzheimer's Disease Neuroimaging Initiative . Longitudinal pathways of cerebrospinal fluid and positron emission tomography biomarkers of amyloid‐β positivity. Mol Psychiatry. 2021;26(10):5864‐5874. [DOI] [PMC free article] [PubMed] [Google Scholar]
8. Morenas‐Rodríguez E, Li Y, Nuscher B, et al. Soluble TREM2 in CSF and its association with other biomarkers and cognition in autosomal‐dominant Alzheimer's disease: a longitudinal observational study. Lancet Neurol. 2022;21(4):329‐341. [DOI] [PMC free article] [PubMed] [Google Scholar]
9. Ewers M, Franzmeier N, Suárez‐Calvet M, et al.; Alzheimer's Disease Neuroimaging Initiative . Increased soluble TREM2 in cerebrospinal fluid is associated with reduced cognitive and clinical decline in Alzheimer's disease. Sci Transl Med. 2019;11(507):eaav6221. [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Ewers M, Biechele G, Suárez‐Calvet M, et al. Higher CSF sTREM2 and microglia activation are associated with slower rates of beta‐amyloid accumulation. EMBO molecular medicine. 2020;12(9):e12308. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Pereira JB, Janelidze S, Strandberg O, et al. Microglial activation protects against accumulation of tau aggregates in nondemented individuals with underlying Alzheimer's disease pathology. Nat Aging. 2022:1‐7. [DOI] [PMC free article] [PubMed] [Google Scholar]
12. Qu W, Li L. Microglial TREM2 at the intersection of brain aging and Alzheimer's disease. Neuroscientist. 2023;29(3):302‐316. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Brown GC, George‐Hyslop S. Does soluble TREM2 protect against Alzheimer's disease?. Front Aging Neurosci. 2022;13:834697. [DOI] [PMC free article] [PubMed] [Google Scholar]
14. Biel D, Suárez‐Calvet M, Hager P, et al.; Alzheimer's Disease Neuroimaging Initiative . sTREM2 is associated with amyloid‐related p‐tau increases and glucose hypermetabolism in Alzheimer's disease. EMBO Mol Med. 2023;15:e16987. [DOI] [PMC free article] [PubMed] [Google Scholar]
15. Goukasian N, Hwang KS, Romero T, et al. Association of brain amyloidosis with the incidence and frequency of neuropsychiatric symptoms in ADNI: a multisite observational cohort study. BMJ Open. 2019;9(12):e031947. doi: 10.1136/bmjopen-2019-031947 [DOI] [PMC free article] [PubMed] [Google Scholar]
16. Kleinberger G, Yamanishi Y, Suarez‐Calvet M, et al. TREM2 mutations implicated in neurodegeneration impair cell surface transport and phagocytosis. Sci Transl Med. 2014;6(243):243ra86. doi: 10.1126/scitranslmed.3009093 [DOI] [PubMed] [Google Scholar]
17. Suárez‐Calvet M, Kleinberger G, MÁ AraqueCaballero, et al. sTREM 2 cerebrospinal fluid levels are a potential biomarker for microglia activity in early‐stage Alzheimer's disease and associate with neuronal injury markers. EMBO Mol Med. 2016;8(5):466‐476. [DOI] [PMC free article] [PubMed] [Google Scholar]
18. Landau SM, Mintun MA, Joshi AD, et al. Amyloid deposition, hypometabolism, and longitudinal cognitive decline. Ann Neurol. 2012;72(4):578‐586. doi: 10.1002/ana.23650 [DOI] [PMC free article] [PubMed] [Google Scholar]
19. Landau SM, Lu M, Joshi AD, et al.; Alzheimer's Disease Neuroimaging Initiative . Comparing positron emission tomography imaging and cerebrospinal fluid measurements of β‐amyloid. Ann Neurol. 2013;74(6):826‐836. [DOI] [PMC free article] [PubMed] [Google Scholar]
20. Joshi AD, Pontecorvo MJ, Clark CM, et al.; Florbetapir F 18 Study Investigators . Performance characteristics of amyloid PET with florbetapir F 18 in patients with Alzheimer's disease and cognitively normal subjects. J Nucl Med. 2012;53(3):378‐384. [DOI] [PubMed] [Google Scholar]
21. Hansson O, Seibyl J, Stomrud E, et al. CSF biomarkers of Alzheimer's disease concord with amyloid‐β PET and predict clinical progression: a study of fully automated immunoassays in BioFINDER and ADNI cohorts. Alzheimers Dement. 2018;14(11):1470‐1481. [DOI] [PMC free article] [PubMed] [Google Scholar]
22. Hansson O, Lehmann S, Otto M, Zetterberg H, Lewczuk P. Advantages and disadvantages of the use of the CSF Amyloid β (Aβ) 42/40 ratio in the diagnosis of Alzheimer's Disease. Alzheimers Res Ther. 2019;11:1‐15. [DOI] [PMC free article] [PubMed] [Google Scholar]
23. Santangelo R, Cecchetti G, Bernasconi MP, et al. Cerebrospinal fluid amyloid‐β 42, total tau and phosphorylated tau are low in patients with normal pressure hydrocephalus: analogies and differences with Alzheimer's disease. J Alzheimers Dis. 2017;60(1):183‐200. [DOI] [PubMed] [Google Scholar]
24. Kapaki E, Paraskevas G, Tzerakis N, et al. Cerebrospinal fluid tau, phospho‐tau181 and β‐amyloid1− 42 in idiopathic normal pressure hydrocephalus: a discrimination from Alzheimer's disease. Eur J Neurol. 2007;14(2):168‐173. [DOI] [PubMed] [Google Scholar]
25. Graff‐Radford J, Jones DT, Wiste HJ, et al. Cerebrospinal fluid dynamics and discordant amyloid biomarkers. Neurobiol Aging. 2022;110:27‐36. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Mori E, Ishikawa M, Kato T, et al. Guidelines for management of idiopathic normal pressure hydrocephalus. Neurologia Med Chir. 2012;52(11):775‐809. [DOI] [PubMed] [Google Scholar]
27. Suárez‐Calvet M, Morenas‐Rodríguez E, Kleinberger G, et al.; Alzheimer's Disease Neuroimaging Initiative . Early increase of CSF sTREM2 in Alzheimer's disease is associated with tau related‐neurodegeneration but not with amyloid‐β pathology. Mol Neurodegener. 2019;14:1‐14. [DOI] [PMC free article] [PubMed] [Google Scholar]
28. Graff‐Radford J, Gunter JL, Jones DT, et al. Cerebrospinal fluid dynamics disorders: relationship to Alzheimer biomarkers and cognition. Neurology. 2019;93(24):e2237‐e2246. [DOI] [PMC free article] [PubMed] [Google Scholar]
29. Dong M‐H, Zhou L‐Q, Tang Y, et al. CSF sTREM2 in neurological diseases: a two‐sample Mendelian randomization study. J Neuroinflammation. 2022;19(1):79. [DOI] [PMC free article] [PubMed] [Google Scholar]
30. Liu C, Yu J. Genome‐Wide Association Studies for cerebrospinal fluid soluble TREM2 in Alzheimer's disease. Front Aging Neurosci. 2019;11:297. doi: 10.3389/fnagi.2019.00297 [DOI] [PMC free article] [PubMed] [Google Scholar]
31. Tang W, Huang Q, Yao Y‐Y, Wang Y, Wu Y‐L, Wang Z‐Y. Does CSF p‐tau 181 help to discriminate Alzheimer's disease from other dementias and mild cognitive impairment? A meta‐analysis of the literature. J Neural Transm. 2014;121:1541‐1553. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information

DAD2-17-e70072-s002.docx^{(515.9KB, docx)}

Supporting Information

DAD2-17-e70072-s001.pdf^{(1.2MB, pdf)}

Data Availability Statement

All raw data used in this study are freely available from https://ida.loni.usc.edu.

[dad270072-bib-0001] 1. Jack CR, Jr ., Bennett DA, Blennow K, et al. NIA‐AA Research Framework: toward a biological definition of Alzheimer's disease. Alzheimers Dement. 2018;14(4):535‐562. doi: 10.1016/j.jalz.2018.02.018 [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0002] 2. Jia J, Ning Y, Chen M, et al. Biomarker changes during 20 years preceding Alzheimer's disease. N Eng J Med. 2024;390:712‐722. [DOI] [PubMed] [Google Scholar]

[dad270072-bib-0003] 3. Leitão MJ, Silva‐Spínola A, Santana I, et al. Clinical validation of the Lumipulse G cerebrospinal fluid assays for routine diagnosis of Alzheimer's disease. Alzheimer's Research & Therapy. 2019;11:1‐12. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0004] 4. Campbell MR, Ashrafzadeh‐Kian S, Petersen RC, et al. P‐tau/aβ42 and aβ42/40 ratios in csf are equally predictive of amyloid pet status. Alzheimers Dement. 2021;13(1):e12190. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0005] 5. De Wilde A, Reimand J, Teunissen CE, et al. Discordant amyloid‐β PET and CSF biomarkers and its clinical consequences. Alzheimers Res Ther. 2019;11(1):1‐13. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0006] 6. Vos SJ, Gordon BA, Su Y, et al. NIA‐AA staging of preclinical Alzheimer disease: discordance and concordance of CSF and imaging biomarkers. Neurobiol Aging. 2016;44:1‐8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0007] 7. Sala A, Nordberg A, Rodriguez‐Vieitez E; Alzheimer's Disease Neuroimaging Initiative . Longitudinal pathways of cerebrospinal fluid and positron emission tomography biomarkers of amyloid‐β positivity. Mol Psychiatry. 2021;26(10):5864‐5874. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0008] 8. Morenas‐Rodríguez E, Li Y, Nuscher B, et al. Soluble TREM2 in CSF and its association with other biomarkers and cognition in autosomal‐dominant Alzheimer's disease: a longitudinal observational study. Lancet Neurol. 2022;21(4):329‐341. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0009] 9. Ewers M, Franzmeier N, Suárez‐Calvet M, et al.; Alzheimer's Disease Neuroimaging Initiative . Increased soluble TREM2 in cerebrospinal fluid is associated with reduced cognitive and clinical decline in Alzheimer's disease. Sci Transl Med. 2019;11(507):eaav6221. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0010] 10. Ewers M, Biechele G, Suárez‐Calvet M, et al. Higher CSF sTREM2 and microglia activation are associated with slower rates of beta‐amyloid accumulation. EMBO molecular medicine. 2020;12(9):e12308. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0011] 11. Pereira JB, Janelidze S, Strandberg O, et al. Microglial activation protects against accumulation of tau aggregates in nondemented individuals with underlying Alzheimer's disease pathology. Nat Aging. 2022:1‐7. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0012] 12. Qu W, Li L. Microglial TREM2 at the intersection of brain aging and Alzheimer's disease. Neuroscientist. 2023;29(3):302‐316. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0013] 13. Brown GC, George‐Hyslop S. Does soluble TREM2 protect against Alzheimer's disease?. Front Aging Neurosci. 2022;13:834697. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0014] 14. Biel D, Suárez‐Calvet M, Hager P, et al.; Alzheimer's Disease Neuroimaging Initiative . sTREM2 is associated with amyloid‐related p‐tau increases and glucose hypermetabolism in Alzheimer's disease. EMBO Mol Med. 2023;15:e16987. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0015] 15. Goukasian N, Hwang KS, Romero T, et al. Association of brain amyloidosis with the incidence and frequency of neuropsychiatric symptoms in ADNI: a multisite observational cohort study. BMJ Open. 2019;9(12):e031947. doi: 10.1136/bmjopen-2019-031947 [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0016] 16. Kleinberger G, Yamanishi Y, Suarez‐Calvet M, et al. TREM2 mutations implicated in neurodegeneration impair cell surface transport and phagocytosis. Sci Transl Med. 2014;6(243):243ra86. doi: 10.1126/scitranslmed.3009093 [DOI] [PubMed] [Google Scholar]

[dad270072-bib-0017] 17. Suárez‐Calvet M, Kleinberger G, MÁ AraqueCaballero, et al. sTREM 2 cerebrospinal fluid levels are a potential biomarker for microglia activity in early‐stage Alzheimer's disease and associate with neuronal injury markers. EMBO Mol Med. 2016;8(5):466‐476. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0018] 18. Landau SM, Mintun MA, Joshi AD, et al. Amyloid deposition, hypometabolism, and longitudinal cognitive decline. Ann Neurol. 2012;72(4):578‐586. doi: 10.1002/ana.23650 [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0019] 19. Landau SM, Lu M, Joshi AD, et al.; Alzheimer's Disease Neuroimaging Initiative . Comparing positron emission tomography imaging and cerebrospinal fluid measurements of β‐amyloid. Ann Neurol. 2013;74(6):826‐836. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0020] 20. Joshi AD, Pontecorvo MJ, Clark CM, et al.; Florbetapir F 18 Study Investigators . Performance characteristics of amyloid PET with florbetapir F 18 in patients with Alzheimer's disease and cognitively normal subjects. J Nucl Med. 2012;53(3):378‐384. [DOI] [PubMed] [Google Scholar]

[dad270072-bib-0021] 21. Hansson O, Seibyl J, Stomrud E, et al. CSF biomarkers of Alzheimer's disease concord with amyloid‐β PET and predict clinical progression: a study of fully automated immunoassays in BioFINDER and ADNI cohorts. Alzheimers Dement. 2018;14(11):1470‐1481. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0022] 22. Hansson O, Lehmann S, Otto M, Zetterberg H, Lewczuk P. Advantages and disadvantages of the use of the CSF Amyloid β (Aβ) 42/40 ratio in the diagnosis of Alzheimer's Disease. Alzheimers Res Ther. 2019;11:1‐15. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0023] 23. Santangelo R, Cecchetti G, Bernasconi MP, et al. Cerebrospinal fluid amyloid‐β 42, total tau and phosphorylated tau are low in patients with normal pressure hydrocephalus: analogies and differences with Alzheimer's disease. J Alzheimers Dis. 2017;60(1):183‐200. [DOI] [PubMed] [Google Scholar]

[dad270072-bib-0024] 24. Kapaki E, Paraskevas G, Tzerakis N, et al. Cerebrospinal fluid tau, phospho‐tau181 and β‐amyloid1− 42 in idiopathic normal pressure hydrocephalus: a discrimination from Alzheimer's disease. Eur J Neurol. 2007;14(2):168‐173. [DOI] [PubMed] [Google Scholar]

[dad270072-bib-0025] 25. Graff‐Radford J, Jones DT, Wiste HJ, et al. Cerebrospinal fluid dynamics and discordant amyloid biomarkers. Neurobiol Aging. 2022;110:27‐36. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0026] 26. Mori E, Ishikawa M, Kato T, et al. Guidelines for management of idiopathic normal pressure hydrocephalus. Neurologia Med Chir. 2012;52(11):775‐809. [DOI] [PubMed] [Google Scholar]

[dad270072-bib-0027] 27. Suárez‐Calvet M, Morenas‐Rodríguez E, Kleinberger G, et al.; Alzheimer's Disease Neuroimaging Initiative . Early increase of CSF sTREM2 in Alzheimer's disease is associated with tau related‐neurodegeneration but not with amyloid‐β pathology. Mol Neurodegener. 2019;14:1‐14. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0028] 28. Graff‐Radford J, Gunter JL, Jones DT, et al. Cerebrospinal fluid dynamics disorders: relationship to Alzheimer biomarkers and cognition. Neurology. 2019;93(24):e2237‐e2246. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0029] 29. Dong M‐H, Zhou L‐Q, Tang Y, et al. CSF sTREM2 in neurological diseases: a two‐sample Mendelian randomization study. J Neuroinflammation. 2022;19(1):79. [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0030] 30. Liu C, Yu J. Genome‐Wide Association Studies for cerebrospinal fluid soluble TREM2 in Alzheimer's disease. Front Aging Neurosci. 2019;11:297. doi: 10.3389/fnagi.2019.00297 [DOI] [PMC free article] [PubMed] [Google Scholar]

[dad270072-bib-0031] 31. Tang W, Huang Q, Yao Y‐Y, Wang Y, Wu Y‐L, Wang Z‐Y. Does CSF p‐tau 181 help to discriminate Alzheimer's disease from other dementias and mild cognitive impairment? A meta‐analysis of the literature. J Neural Transm. 2014;121:1541‐1553. [DOI] [PubMed] [Google Scholar]

PERMALINK

sTREM2 in discordant CSF Aβ42 and p‐tau181

Danni Li

William G Mantyh

Lu Men

Ishika Jain

Matthew Glittenberg

Binchong An

Lin Zhang

Ling Li

Abstract

INTRODUCTION

METHODS

RESULTS

DISCUSSION

Highlights

1. BACKGROUND

2. METHODS

2.1. Study participants

2.2. AD biomarkers

RESEARCH CONTEXT

2.3. MRI scans

2.4. Statistical analysis

TABLE 1.

3. RESULTS

3.1. Categorization of study participants based on CSF Aβ42 or p‐tau181/Aβ42 and Aβ PET

3.2. sTREM2 in participants with discordant CSF Aβ42 and p‐tau181/Aβ42

TABLE 2.

FIGURE 1.

FIGURE 2.

3.3. sTREM2 in participants with discordant CSF Aβ42 and Aβ PET

FIGURE 3.

4. DISCUSSION

5. CONCLUSIONS

CONFLICT OF INTEREST STATEMENT

CONSENT STATEMENT

Supporting information

ACKNOWLEDGMENTS

DATA AVAILABILITY STATEMENT

REFERENCES

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases

sTREM2 in discordant CSF Aβ₄₂ and p‐tau181

3.2. sTREM2 in participants with discordant CSF Aβ₄₂ and p‐tau181/Aβ₄₂

3.3. sTREM2 in participants with discordant CSF Aβ₄₂ and Aβ PET