Amyloid marker levels and the risk of developing cerebral small vessel disease: A Mendelian randomization study

Abstract

Background

Previous observational studies have suggested a potential association between amyloid marker levels and the risk of developing cerebral small vessel disease (CSVD), but this relationship remains incompletely understood.

Objective

This study was conducted to assess the impact of amyloid marker levels on the risk of developing CSVD via Mendelian randomization (MR) design.

Methods

Using the latest genome-wide association study summary statistics for 5 plasma amyloid markers and 4 CSVD traits, a two-sample MR study was conducted to assess the genetic relationship between amyloid marker levels and CSVD risk. Furthermore, reverse MR analysis was utilized to establish the causal relationship between CSVD traits and the levels of the identified plasma amyloid markers to explore potential bidirectional causality.

Results

After FDR correction, greater amyloid-β (Aβ) 42 levels were associated with an increased risk of developing lobar cerebral microbleeds (CMBs) (odds ratio = 2.311, 95% confidence interval 1.403–3.809, p = 0.001, p FDR = 0.040). Potential positive correlations were detected between Aβ₄₀ levels and the risk of developing intracerebral hemorrhage, between Aβ₄₂ levels and the risk of developing all CMBs, and between serum amyloid P component levels and white matter fractional anisotropy status. In reverse MR analysis, no effect of CSVD traits on amyloid marker levels was detected.

Conclusions

Our study suggests potential causal relationships between amyloid marker levels and different CSVD traits. Our results contribute to a greater understanding of the pathophysiology of CSVD, particularly in relation to amyloid deposition.

Keywords

Alzheimer’s disease amyloid marker cerebral small vessel disease genetics instrumental variables Mendelian randomization

Introduction

Cerebral small vessel disease (CSVD) is a series of clinical, imaging and pathologic syndromes caused by lesions in intracranial small arteries, arterioles, capillaries, venules, and small veins.¹ Clinical manifestations include both acute manifestations of small vessel stroke (SVS) and intracerebral hemorrhaged (ICH) and chronic manifestations of cognitive, motor, affective, and diaphoretic disorders.² Imaging features commonly associated with CSVD include white matter hyperintensities (WMHs) and cerebral microbleeds (CMBs).³ Epidemiological studies indicate that CSVD has imposed a significant health burden, particularly in low- and middle-income countries, in recent years.⁴ Despite its prevalence, the pathophysiologic cause of CSVD is currently unknown. Hypertension and amyloid deposition are thought to be the main causes.^5–8

The amyloid deposition hypothesis has gained increasing attention in CSVD research.⁴ Amyloid-β protein precursor protein (AβPP) is hydrolyzed to amyloid-β (Aβ) by β- and γ-secretase enzymes. Then, Aβ combines with serum amyloid P component (SAP) to form amyloid plaques. Ultimately, it is deposited in small- and medium-sized arteries, causing a series of pathophysiological changes.^9–12 Traditional methods for detecting amyloid include autopsy, positron emission tomography (PET), and cerebrospinal fluid (CSF) analysis.¹³ However, owing to the high cost of PET and the invasive nature of CSF collection, plasma amyloid testing has become a more attractive method of mass screening.¹⁴ Several studies have shown a role for plasma Aβ in CSVD. However, the results of previous observational studies are conflicting. Some research has shown increased plasma Aβ levels in patients with CSVD compared with those in normal control participants.^15–21 Other studies have reported no significant differences in plasma Aβ levels among patients with CSVD.²² Reverse causality interference, confounding variables, and varying sample sizes could be the causes of these contradictory results. Importantly, these studies do not clarify whether plasma amyloid is a cause or a consequence of CSVD development.

Randomized controlled trials (RCTs) have the potential to address the drawbacks of observational research. However, RCTs have high economic and time costs and may not be conducted strictly. Mendelian randomization (MR) can be used to establish a relationship between exposure and outcome by using genetic variants as instrumental variables (IVs).²³ Since genetic alleles are randomly assigned during conception, the confounding bias and reverse causal associations present in observational studies are avoided.²⁴ Therefore, MR enables investigators to draw conclusions about causality, particularly when RCTs are not possible.

To investigate the relationships between plasma amyloid marker levels (AβPP, amyloid precursor-like protein 2 (APLP2), SAP, and Aβ peptide levels) and CSVD traits (ICH, SVS, CMBs, and WMH), we performed a two-sample MR study. The most recent and extensive genome-wide association study (GWAS) datasets were utilized in our analysis. It is worth emphasizing that biomarkers for CSVD encompass multiple domains including imaging, body fluids, and genetics.²⁵ This study aims to investigate the causal relationship between amyloid deposition and CSVD, thus focusing on the amyloid pathway. Furthermore, limiting the analysis to biologically coherent amyloid pathways helps ensure the robustness of causal inference and enhances the clarity of mechanistic explanations.

Methods

Data source

Genetic predictors of APP, APLP2, and SAP were derived from a recent human plasma proteomic GWAS, which included data on more than 3600 proteins from 3301 individuals of European ancestry from the INTERVAL study. The mean age of the study population was 43.7 years, and 51.1% of the participants were male. Genotyping and genotype imputation were performed in this study, and plasma protein levels were quantitatively measured using the SOMAscan platform. Genome-wide association analyses were then conducted to identify genetic variants associated with protein levels.²⁶ To ensure the reliability of results, we supplemented genetic variants for APP and APLP from the most recently published databases. Genetic predictors were derived from a large-scale GWAS of human plasma proteomics, which included data from 35,559 Icelandic participants. The study population had a mean age of 55 years (standard deviation = 17 years), with 57% being female. Plasma protein levels were quantified using the SomaScan 4.0 platform. A GWAS was then conducted to examine associations between 27.2 million sequence variants and plasma protein levels, ultimately identifying protein-associated quantitative trait loci (PQTLs) where sequence variants showed significant correlations with protein levels.²⁷ The three-city (3C) study, Rotterdam study, Pittsburgh Cardiovascular Health Study–Cognition Study (CHS-CS), and Alzheimer's Disease Neuroimaging Initiative (ADNI) study, which included a total of 3528 healthy people of European descent with a mean age of 71.5 years and a male proportion of 59.8%, provided genetic predictors of plasma Aβ₄₀ and Aβ₄₂ levels. These predictors were derived through a standardized genotyping and imputation pipeline, harmonized Aβ quantification procedures, and fixed-effects meta-analysis, resulting in robust genetic instruments²⁸ (Figure 1A).

Figure 1.

Overview of the study design. (A) Description of the amyloid markers used for exposure. (B) Description of the cerebral small-vessel disease traits used as outcomes. (C) MR framework. APLP2: amyloid precursor-like protein 2; APP: amyloid-β protein precursor; Aβ: amyloid-β; SAP: serum amyloid P component; GWAS: genome-wide association study; SNP: single-nucleotide polymorphism; ICH: intracerebral hemorrhage; SVS: small vessel stroke; CMB: cerebral microbleeds; WMH: white matter hyperintensities; IVW: inverse-variance weighting; MR: Mendelian randomization.

The CSVD data included genetic variants associated with ICH, SVS CMBs, and WMH. The GWAS data from three European biobanks, which included data from 711,818 normal control participants and 7605 ICH patients, were meta-analyzed to produce the ICH data. The ICH phenotype was defined by clinical diagnosis and confirmed by acute neuroimaging, with secondary hemorrhages excluded. Genotypes were obtained through SNP arrays, imputed, and rigorously quality-controlled. Within each cohort, association analyses were conducted using logistic regression or mixed models, and results were subsequently combined via fixed-effects meta-analysis.²⁹ The SVS data were derived from a meta-analysis published by the GIGASTROKE consortium. SVS was discovered in 13,620 of the 1,241,619 participants. For SVS, each study applied standardized case definitions and quality control procedures. After performing GWAS, robust genetic loci were identified through meta-analysis and cross-population validation.³⁰ Eleven population-based cohort studies and three case‒control or case‒only stroke cohorts that were included in a GWAS provided the data for CMB. Among the 24,354 individuals, 3172 had CMBs: 1932 were lobar, and 1240 were deep. Phenotypes were determined using MRI GRE/SWI sequences, distinguishing lobar and deep CMB types. Genotype data were processed through arrays, imputation, and quality control, and association analyses followed by meta-analysis revealed significant signals in the APOE region.³¹ GWAS WMH data were obtained from the UK Biobank, the Cohorts for Heart and Ageing Research in Genomic Epidemiology (CHARGE) consortium, and a meta-analysis of WMH studies in stroke patients, including 42310 individuals. GWAS data on the mean diffusivity (MD) and mean fractional anisotropy (FA) of 21 main white matter bundles were utilized to indicate the integrity of the microstructure of white matter. Phenotypes were uniformly extracted and quality-controlled following the ENIGMA-DTI protocol, and robust genetic loci were ultimately identified through meta-analysis³² (Figure 1B, Supplemental Table 1).

Selection of genetic predictors

Single-nucleotide polymorphisms (SNPs) that met the genome-wide significance threshold of p < 1× 10⁻⁵ were selected as potential IVs. We tested their independence by performing a linkage disequilibrium (LD) analysis (r² < 0.001, window size = 10,000 kb). Furthermore, the strength of each SNP was evaluated via the F statistic. When the F statistic is greater than 10, the likelihood of weak tool bias is reduced.³³ F = beta²/SE² was the formula used to determine the F statistic for each variant.³⁴ The characteristics of the IVs are summarized in Supplemental Table 2. In addition, to increase the strength of the instrumental variables, we further selected SNPs reaching the genome-wide significance threshold (p < 5 × 10⁻⁸) for analysis. When only one SNP was available under this stricter threshold, the Wald ratio method was applied.

Study design

We used a bidirectional two-sample MR study to analyze causal relationships between amyloid markers and CSVD and vice versa. MR analyses must satisfy the following 3 hypotheses. Hypothesis 1: There is a significant association between IVs and exposure factors. Hypothesis 2: There are no associations between IVs and any confounders affecting outcome and exposure. Hypothesis 3: Exposure is the only route by which IVs affect outcome.³⁵ (Figure 1C) This study was conducted in adherence to the STROBE-MR guidelines as shown in Supplemental Material.

MR analysis

The principles and methods of MR have been described in previous studies.^36–38 In the present study, the main method used to analyze causality was the inverse-variance weighting (IVW) method. Simple median, weighted median, MR‒Egger, and pleiotropy residual sum and outlier (MR–PRESSO) were used as complementary methods.^39–41 We tested for heterogeneity via Cochran's Q test.⁴² For exposures where there was heterogeneity, we used a random-effects IVW model; otherwise, we used a fixed-effects IVW model. The MR–PRESSO and MR‒Egger intercepts were used to detect and assess horizontal pleiotropic outliers. The impact of each individual SNP on overall causation was independently evaluated via the leave-one-out (LOO) approach.

To conduct our analyses, we used the “TwoSampleMR” package in R version 4.0.3. p < 0.05 was considered to indicate statistical significance. We corrected the p values using the false discovery rate (FDR).

Results

Associations of plasma aβPP, APLP2, SAP, and Aβ peptide levels with the risk of developing CSVD

Before performing all analyses, we detected heterogeneity in the IV effects of plasma Aβ₄₂ with ICH, all CMBs, deep CMBs, WMH and WM MD (ICH: Cochran's Q = 54.240, p < 0.001; all CMBs: Cochran's Q = 27.489, p < 0.001; deep CMBs: Cochran's Q = 14.229, p = 0.02; WMH: Cochran's Q = 43.391, p = 0.001; WM MD: Cochran's Q = 30.246, p = 0.035), AβPP and WM FA (Cochran's Q = 34.119, p = 0.048), and SAP and WM MD (Cochran's Q = 26.298, p = 0.035) (Table 1). Therefore, for those analyses, we used the random-effects IVW model. The other analyses yielded no heterogeneity and were thus assessed via fixed-effects IVW models.

Table 1.

Summary of the additional MR analysis for the effect of amyloid marker levels on the risk of developing CSVD.

		Cochran's test			MR-Egger test		MR PRESSO
Exposure	Outcome	Q	Q_df	Q_p	Intercept	p	global test (p)	outlier-corrected
APP	ICH	20.023	21	0.52	0.001	0.954	0.501	NA
APLP2	ICH	11.31	16	0.79	0.026	0.122	0.779	NA
SAP	ICH	9.129	16	0.908	−0.021	0.19	0.899	NA
Aβ40	ICH	21.998	34	0.944	0	0.949	0.957	NA
Aβ42	ICH	54.24	17	0	0.012	0.547	0.006	0.106
APP	SVS	14.317	12	0.281	−0.004	0.922	0.285	NA
APLP2	SVS	5.189	8	0.737	0.002	0.931	0.752	NA
SAP	SVS	12.559	10	0.249	0.017	0.465	0.270	NA
Aβ40	SVS	27.388	18	0.072	−0.034	0.098	0.076	NA
Aβ42	SVS	2.620	5	0.758	−0.016	0.745	0.783	NA
APP	All CMB	9.299	16	0.901	−0.001	0.981	0.902	NA
APLP2	All CMB	5.648	13	0.958	0.039	0.218	0.948	NA
SAP	All CMB	13.233	11	0.278	−0.012	0.705	0.323	NA
Aβ40	All CMB	18.768	19	0.472	0.001	0.982	0.474	NA
Aβ42	All CMB	27.489	7	0	−0.046	0.574	0.005	0.374
APP	Deep CMB	14.66	11	0.199	−0.006	0.963	0.187	NA
APLP2	Deep CMB	5.063	9	0.829	0.021	0.708	0.813	NA
SAP	Deep CMB	5.844	8	0.665	−0.025	0.734	0.698	NA
Aβ40	Deep CMB	13.459	16	0.639	0.027	0.587	0.653	NA
Aβ42	Deep CMB	14.229	6	0.027	−0.116	0.26	0.044	0.256
APP	Lobar CMB	8.059	14	0.886	−0.053	0.369	0.89	NA
APLP2	Lobar CMB	7.444	11	0.762	0.073	0.102	0.768	NA
SAP	Lobar CMB	16.358	11	0.128	−0.016	0.741	0.155	NA
Aβ40	Lobar CMB	18.566	17	0.354	−0.048	0.2	0.337	NA
Aβ42	Lobar CMB	10.844	6	0.093	−0.095	0.171	0.12	NA
APP	WMH	23.486	22	0.375	0.011	0.206	0.408	NA
APLP2	WMH	9.761	16	0.879	−0.003	0.711	0.87	NA
SAP	WMH	14.201	15	0.51	−0.004	0.663	0.557	NA
Aβ40	WMH	27.449	34	0.779	0.003	0.331	0.829	NA
Aβ42	WMH	43.391	18	0.001	−0.004	0.647	0.006	0.315
APP	WM FA	34.119	22	0.048	−0.027	0.545	0.05	NA
APLP2	WM FA	21.499	16	0.16	0.046	0.286	0.188	NA
SAP	WM FA	15.377	15	0.425	0.004	0.908	0.486	NA
Aβ40	WM FA	19.858	34	0.974	0.003	0.836	0.974	NA
Aβ42	WM FA	21.084	18	0.275	−0.029	0.22	0.359	NA
APP	WM MD	32.306	22	0.072	0.059	0.18	0.078	NA
APLP2	WM MD	23.63	16	0.098	−0.048	0.303	0.108	NA
SAP	WM MD	26.298	15	0.035	0.007	0.895	0.047	0.235
Aβ40	WM MD	30.792	34	0.626	0.011	0.448	0.58	NA
Aβ42	WM MD	30.246	18	0.035	0.051	0.074	0.071	NA

APP: amyloid-β protein precursor; APLP2: amyloid precursor-like protein 2; SAP: serum amyloid P component; Aβ: amyloid beta; ICH: intracerebral hemorrhage; SVS: small vessel stroke; CMB: cerebral microbleeds; WMH: white matter hyperintensities; FA: fractional anisotropy; MD: mean diffusivity

High plasma Aβ₄₂ levels predicted by genetics were associated with a greater risk of developing lobar CMBs (IVW odds ratio (OR) = 2.311, 95% confidence interval (CI) 1.403–3.809, p = 0.001) (Figure 2). This result remained stable after FDR correction (p = 0.040). All sensitivity analyses yielded consistent results (weighted median, OR = 2.005, 95% CI 1.093–3.677, p = 0.025; MR–Egger, OR = 5.953, 95% CI 1.719–20.620, p = 0.037; simple median: OR = 1.941, 95% CI 1.027–3.666, p = 0.041; MR–PRESSO: p = 0.017) (Figure 3A). The heterogeneity test revealed no significant heterogeneity among the SNPs (Cochran's Q = 10.844, p = 0.093). The MR–PRESSO method showed no outliers in the IVs. The MR–Egger intercept (−0.095, p = 0.171) showed no evidence of directional pleiotropy, suggesting that the results are robust (Supplemental Table 3).

Figure 2.

Causal effects of amyloid marker levels on the risk of developing CSVD identified via inverse-variance weighting. The red dots represent the OR value, and the straight line represents the 95% CI. SNP: single-nucleotide polymorphism; OR: odds ratio; 95% CI, 95% confidence interval; FDR: false discovery rate; APLP2: amyloid precursor-like protein 2; APP: amyloid-β protein precursor; Aβ: amyloid-β; SAP: serum amyloid P component; ICH: intracerebral hemorrhage; SVS: small vessel stroke; CMB: cerebral microbleeds; WMH: white matter hyperintensities; FA: fractional anisotropy; MD: mean diffusivity.

Figure 3.

Scatterplot of the effects of amyloid marker levels on the risk of developing cerebral small vessel disease. The slope of each line corresponds to the estimated MR effect of each method. (A) Aβ₄₂ and lobar cerebral microbleeds. (B) Aβ₄₂ and all cerebral microbleeds. (C) Aβ₄₀ and intracerebral hemorrhage. (D) Serum amyloid P component and white matter fractional anisotropy. IVW: inverse-variance weighting; MR: Mendelian randomization.

According to our findings, genetically predicted increased plasma Aβ₄₂ levels were associated with an increased risk of developing all CMBs (IVW, OR = 1.920, 95% CI 1.096–3.362, p = 0.023) (Figure 2), and sensitivity analyses yielded directionally consistent results (Figure 3B). There was no multiplicity of IVs (MR–Egger's intercept = 0.046, p = 0.574), and the MR–PRESSO analysis revealed no outliers (Supplemental Table 3).

There was a potential positive correlation between genetically predicted increased plasma Aβ₄₀ levels and the risk of developing ICH. The IVW analysis revealed that for every one-unit increase in genetically predicted Aβ₄₀, there was a 4% greater probability of ICH occurrence (OR = 1.040, 95% CI 1.003–1.079, p = 0.032) (Figure 2). However, the sensitivity analyses performed via MR–PRESSO yielded significantly different results (p = 0.012). Other sensitivity analyses yielded directionally consistent results (Figure 3C). The MR–Egger intercept did not reveal pleiotropy in the IVs (MR–Egger's intercept = −0.0004, p = 0.949) (Supplemental Table 3).

Genetically predicted increased SAP levels were also a potential risk factor for WM FA (IVW, OR = 1.260, 95% CI 1.075–1.478, p = 0.004) (Figure 2). The MR–PRESSO results were consistent (p = 0.012), and the sensitivity analyses yielded the same trend (Figure 3D). No pleiotropy or heterogeneity was detected (MR–Egger's intercept = 0.004, p = 0.908) (Supplemental Table 3).

In summary, our research indicates that gene-mediated changes in plasma Aβ₄₂, Aβ₄₀, and SAP levels are potential risk factors for CSVD development. However, we do not have sufficient evidence to suggest a causal relationship between plasma AβPP or APLP2 levels and the risk of developing CSVD (Figure 2).

In addition, we further selected instrumental variables using the standard threshold (p < 5 × 10⁻⁸). After FDR correction, genetically predicted higher plasma Aβ₄₂ levels were associated with increased risks of ICH, all CMBs, deep CMBs, lobar CMBs, WMH, and WM MD. No associations were observed between other amyloid markers and CSVD (Supplemental Table 4).

Effects of CSVD status on AβPP, APLP2, SAP, and Aβ peptide levels

Before performing the reverse MR analysis, we also performed a heterogeneity test. We found heterogeneity in the association between all CMBs and SAP levels (Cochran's Q = 40.910, p = 0.023), as well as for the associations of all CMBs, lobar CMBs, and ICH with Aβ₄₂ (all CMBs: Cochran's Q = 70.455, p < 0.001; lobar CMBs: Cochran's Q = 45.951, p < 0.001; ICH: Cochran's Q = 70.704, p < 0.001) (Table 2). Therefore, we used a random-effects IVW model.

Table 2.

Summary of the additional MR analysis for the effect of CSVD traits on amyloid marker levels.

		Cochran's test			MR-Egger test		MR PRESSO
Exposure	Outcome	Q	Q_df	Q_p	Intercept	p	global test (p)	outlier-corrected
ICH	APP	21.443	33	0.939	0.01	0.589	0.938	NA
ICH	APLP2	27.076	33	0.756	0.02	0.286	0.767	NA
ICH	SAP	33.997	33	0.419	−0.009	0.647	0.416	NA
ICH	Abeta40	30.178	33	0.608	−0.02	0.079	0.616	NA
ICH	Abeta42	70.704	33	0	−0.022	0.181	<0.001	0.381
SVS	APP	41.994	40	0.385	−0.034	0.043	0.388	NA
SVS	APLP2	32.783	40	0.784	−0.021	0.202	0.768	NA
SVS	SAP	53.881	40	0.070	−0.026	0.172	0.069	NA
SVS	Abeta40	46.262	40	0.230	−0.005	0.623	0.243	NA
SVS	Abeta42	38.867	40	0.521	0.008	0.427	0.493	NA
All CMB	APP	22.224	25	0.623	0.045	0.053	0.627	NA
All CMB	APLP2	25.931	25	0.411	0.058	0.016	0.432	NA
All CMB	SAP	40.91	25	0.023	0.024	0.412	0.026	0.038
All CMB	Abeta40	15.267	24	0.913	−0.005	0.732	0.922	NA
All CMB	Abeta42	70.455	24	0	−0.027	0.268	<0.001	0.074
Deep CMB	APP	11.125	13	0.6	0.001	0.97	0.621	NA
Deep CMB	APLP2	13.507	13	0.409	−0.028	0.371	0.445	NA
Deep CMB	SAP	15.092	13	0.302	−0.031	0.335	0.343	NA
Deep CMB	Abeta40	8.597	13	0.803	0.028	0.112	0.826	NA
Deep CMB	Abeta42	11.87	13	0.538	−0.01	0.584	0.545	NA
Lobar CMB	APP	14.601	15	0.481	−0.024	0.403	0.503	NA
Lobar CMB	APLP2	19.924	15	0.175	−0.003	0.938	0.154	NA
Lobar CMB	SAP	19.49	15	0.192	−0.054	0.088	0.21	NA
Lobar CMB	Abeta40	8.596	15	0.898	0.003	0.833	0.898	NA
Lobar CMB	Abeta42	45.951	15	0	0.006	0.831	<0.001	0.037
WMH	APP	32.322	43	0.883	0.018	0.237	0.889	NA
WMH	APLP2	38.135	43	0.682	0.008	0.612	0.729	NA
WMH	SAP	39.465	43	0.625	0.019	0.206	0.624	NA
WMH	Abeta40	28.602	42	0.943	0.001	0.916	0.922	NA
WMH	Abeta42	25.133	42	0.982	0.005	0.568	0.984	NA
WM FA	APP	50.759	48	0.365	−0.032	0.019	0.356	NA
WM FA	APLP2	49.384	48	0.418	−0.025	0.06	0.41	NA
WM FA	SAP	43.315	48	0.665	−0.015	0.259	0.646	NA
WM FA	Abeta40	31.108	49	0.978	0.003	0.669	0.977	NA
WM FA	Abeta42	46.324	49	0.582	−0.003	0.705	0.591	NA
WM MD	APP	48.988	53	0.631	−0.014	0.277	0.628	NA
WM MD	APLP2	48.139	53	0.664	−0.013	0.326	0.647	NA
WM MD	SAP	53.201	53	0.466	−0.008	0.522	0.488	NA
WM MD	Abeta40	41.005	49	0.785	−0.008	0.31	0.792	NA
WM MD	Abeta42	50.952	49	0.397	−0.004	0.607	0.412	NA

APP: amyloid-β protein precursor; APLP2: amyloid precursor-like protein 2; SAP: serum amyloid P component; Aβ: amyloid-β; ICH: intracerebral hemorrhage; SVS: small vessel stroke; CMB: cerebral microbleeds; WMH: white matter hyperintensities; FA: fractional anisotropy; MD: mean diffusivity

We used reverse MR analysis with CSVD as the exposure and plasma amyloid marker levels as the outcome to investigate the impact of CSVD traits on these factors. Our analysis revealed that the incidence of all CMBs was associated with increased SAP levels (IVW, OR = 1.116, 95% CI 1.021–1.220, p = 0.016) (Figure 4). However, MR–Egger yielded the opposite results (OR = 0.999, 95% CI 0.760–1.315, p = 0.997). The MR‒Egger intercept was not significant (MR–Egger's intercept = 0.024, p = 0.412), indicating that there was no significant pleiotropy (Supplemental Table 5). This relationship needs to be interpreted with caution. Furthermore, there was no evidence in our research that other CSVD traits and amyloid marker levels were causally related.

Figure 4.

Causal effects of CSVD traits on amyloid density identified by inverse-variance weighting. The red dots represent the OR value, and the straight line represents the 95% CI. SNP: single-nucleotide polymorphism; OR: odds ratio; 95% CI, 95% confidence interval; FDR: false discovery rate; ICH: intracerebral hemorrhage; SVS: small vessel stroke; CMB: cerebral microbleeds; WMH: white matter hyperintensities; FA: fractional anisotropy; MD mean diffusivity; APLP2: amyloid precursor-like protein 2; APP: amyloid-β protein precursor; Aβ: amyloid-β; SAP: serum amyloid P component.

Similarly, when instrumental variables were further selected using the standard threshold (p < 5 × 10⁻⁸), no available instruments were identified for deep CMBs and lobar CMBs; thus, these traits were not included in the analysis. After FDR correction, the results showed that the occurrence of ICH and all CMBs increased plasma Aβ₄₂ levels (Supplemental Table 6).

Discussion

We examined possible causal links between plasma amyloid marker levels and CSVD traits via two-sample MR analysis of genetic IVs from a large-scale GWAS. We discovered that after FDR correction, genetically predicted increased plasma Aβ₄₂ levels were linked to an increased risk of developing lobar CMBs. In addition, there were potential positive causal associations between Aβ₄₂ and all CMBs, between Aβ₄₀ and ICH, and between SAP and WM FA. In reverse MR analysis, no effect of CSVD traits on amyloid marker levels was detected.

This study revealed a significant and stable positive causal relationship between plasma Aβ₄₂ levels and lobar CMBs. Additionally, there were potential associations between plasma Aβ₄₀ and ICH and between Aβ₄₂ and all CMBs. These results suggest that the plasma Aβ peptide plays a harmful role in hemorrhagic lesions. The relationship between Aβ and lobar CMBs has been extensively studied. The most common cause of lobar hemorrhage in elderly individuals is cerebral amyloid angiopathy (CAA). The pathophysiologic mechanism of CAA is the deposition of Aβ in the leptomeningeal and neocortical vessels, which causes lobar hemorrhages and microbleeds. In contrast, hypertensive angiopathy involves the perforator arteries of deep areas first, which causes deep hemorrhages and microbleeds.⁴³ However, the relationship between plasma Aβ levels and the risk of developing CSVD is complex and controversial. Therefore, in the present discussion, we focused on 3 aspects: the relationships of plasma Aβ levels with CSVD imaging phenotypes, ICH, and CAA. (1) The Swedish Bio FINDER study revealed that individuals with CMBs presented increased plasma Aβ₄₂/Aβ₄₀ ratios. In addition, one study revealed elevated Aβ₄₀ levels, but not Aβ₄₂ levels, in patients with CMBs.¹⁶ However, elevated plasma Aβ₄₂ was found to be related to new-onset CMBs in a subsequent longitudinal study.¹⁹ Moreover, several studies have demonstrated that plasma Aβ is linked to other CSVD imaging markers, including WMH and lacunar stroke.^15,17 (2) The association between Aβ₄₀ and ICH in our investigation are comparable with those of a study that indicated that patients with CAA-associated ICH had higher levels of plasma Aβ₄₂, the Aβ₄₂ truncated segment, and Aβ₄₀ than normal control participants did.⁴⁴ Nevertheless, in a different study, there were 42 elderly control subjects, 25 patients with CAA-related bleeding, and 21 patients with bleeding that may have been caused by hypertensive vascular lesions. The investigators did not report any difference in plasma Aβ among the three groups of patients.²² This may be due to the small sample size and lack of assay sensitivity. (3) Piccarducci et al. included 20 patients with CAA and 20 normal control participants. They reported that patients with CAA had higher plasma Aβ₄₀ levels, which explained their lower plasma Aβ_42/40 ratio than that of normal control participants. The inclusion criteria for CAA patients in this study were based on modified Boston criteria, including CMBs or ICH and cortical superficial siderosis.²¹ To some extent, these findings are also consistent with our results. However, Chatterjee et al. reported that plasma Aβ₄₀ levels decreased before symptoms were present in patients with Dutch-type hereditary CAA.⁴⁵ This inconsistency with our findings could be attributed to differences in the study populations. Hereditary CAA may not be fully representative of sporadic CSVD cases. Previous studies had observational designs and could not assess causality between plasma Aβ peptide levels and the risk of developing CSVD. However, our analysis revealed that plasma Aβ peptides are risk factors for the occurrence of CSVD (including ICH and CMBs, especially lobar CMBs) using the largest plasma Aβ peptide and CSVD-related GWAS database available.

The mechanisms underlying the effects of plasma Aβ peptides on hemorrhagic lesions are not fully clear. First, a potential explanation is that the overproduction of plasma Aβ leads to increased accumulation of Aβ in the cerebral arterial media, weakening the vessel wall and leading to hemorrhage.⁴⁶ Second, Aβ may contribute to CAA-associated hemorrhagic lesions not only through direct effects but also by triggering an inflammatory response. Research has shown that amyloids activate perivascular astrocytes and microglia and lead to vascular remodeling, which may be responsible for hemorrhage.⁴⁷ Third, previous studies have reported the presence of CMBs in patients with AD, which might be explained by mixed age-related pathologies.⁴⁸ Interestingly, in our study, different Aβ peptides were associated with distinct events. Aβ₄₂ was a risk factor for CMBs, whereas Aβ₄₀ was linked to ICH. This may be due to the differing properties of these peptides. Aβ₄₂ is less soluble and forms fibrils faster,⁴⁹ so it is more likely to accumulate in capillaries, thereby leading to microbleeds. In contrast, Aβ₄₀ is more soluble, allowing it to diffuse during vascular drainage and deposit in vessel walls, which may lead to ICH.¹¹ Further experimental studies are needed to confirm these underlying mechanisms.

Our study suggests that SAP levels are positively correlated with the incidence of WM FA. SAP is synthesized mainly by hepatocytes, is released into the bloodstream, and can enter the brain through the compromised blood‒brain barrier. Once in the brain, SAP binds to amyloidogenic fibrils composed of Aβ, preventing their hydrolysis and promoting the formation of amyloid plaques.^12,50 However, previous research on SAP and CSVD is limited. Proteomic analyses have shown increased SAP levels in brain tissues from autopsies of patients with CAA.⁵¹ In contrast, no significant changes in serum SAP levels were observed in patients with cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL).⁵² Moreover, the relationship between plasma SAP and AD has been extensively studied. AD and CSVD share commonalities in the amyloidogenic pathways underlying their pathophysiology.⁵³ Elevated SAP levels have been observed in patients with dementia.⁵⁴ Additionally, a recently published MR investigation revealed that plasma SAP is a risk factor for the development of AD.⁵⁵ These results are consistent with the results of our study. To better understand the connection between SAP and the emergence of CSVD, more research is necessary, as the mechanisms behind these two phenomena have not been thoroughly examined.

In this study, we did not find an association between AβPP or APLP2 levels and CSVD traits. Both AβPP and APLP2 are single-pass transmembrane proteins. In vivo, they can be catabolized into Aβ by γ-secretase, which is a rapid and highly complex process.⁵⁶ Previous animal and human studies have shown that mutations in the APP gene are associated with the clinical and imaging manifestations of CSVD.^57,58 The relationship between circulating AβPP and CSVD has not been reported. Interestingly, platelet AβPP has been identified as a diagnostic marker for AD.⁵⁹ The fact that circulating AβPP is present primarily in platelets (>95%) and that plasma free AβPP may not adequately reflect levels in the peripheral circulation could be the reason for the lack of relationship in our investigation. Further investigations into the relationship between platelet AβPP and CSVD are needed to clarify this potential association.

It is important to note that CSVD and AD exhibit multidimensional, complex interactions at the pathophysiological level. One shared feature is Aβ deposition. When Aβ accumulates in the vascular wall, it can lead to the development of CSVD; conversely, its deposition in brain parenchyma forms the characteristic plaques of AD. Imaging studies reveal that WMH and CMBs are common manifestations in both diseases. Genetic factors such as APOE ε4 can increase the risk of both CSVD and AD.^11,60 Mendelian randomization analysis in this study further supports the possibility that the two disorders share amyloid-related pathways. Nevertheless, CSVD and AD exhibit distinct differences in anatomical distribution, biomarker characteristics, and clinical presentation. CSVD predominantly affects cortical and meningeal arterioles, often manifesting as lobar hemorrhages, lobar microbleeds, and superficial cortical iron deposition. In contrast, AD is primarily characterized by parenchymal Aβ deposition, accompanied by diffuse plaque formation and early hippocampal atrophy. Clinically, early AD symptoms predominantly involve memory impairment, whereas CSVD more commonly manifests as stroke, cerebral hemorrhage, or vascular injury ultimately leading to cognitive decline.⁶¹ It is crucial to emphasize that mixed pathology often coexists in the early stages of both conditions, making accurate differentiation based on a single indicator challenging. Only through longitudinal dynamic assessment integrating imaging, CSF biomarkers, and clinical manifestations can precise differentiation be achieved. Therefore, multimodal longitudinal studies will become a key direction for elucidating the interrelationships and underlying mechanisms of these two conditions in the future.

Age is a key factor influencing amyloid levels and CSVD risk. While CSVD-related GWAS data did not provide detailed age information, we systematically reviewed the sources and cohort characteristics of the data used. The ICH GWAS drew from case-control studies or biobanks involving adult and elderly populations.²⁹ The SVS GWAS enrolled stroke patients or high-risk individuals, with peak incidence occurring primarily in those over 60 years of age.³⁰ CMBs GWAS included cohorts such as CHARGE, UK Biobank, and the Rotterdam Study, with average ages predominantly above 55 years.³¹ WMH GWAS samples primarily originated from UK Biobank. Although some analyses integrated pediatric/adolescent cohorts, these constituted a small proportion of the sample size and did not influence the main findings. In summary, existing evidence indicates that the CSVD GWAS data used in this study predominantly represent middle-aged and elderly populations. Furthermore, all original studies adjusted for age as a covariate in their statistical models. Consequently, genetic causal inference is minimally affected by age-related systematic biases.

In the present study, we greatly reduced the interference of confounders and reverse causality by using an MR design. In addition, we used multiple methods to validate the accuracy of our findings and ensure robust effect estimation (F > 10) for each IV. Importantly, we used a large sample size of GWAS data, enhancing the reliability of the results. To our knowledge, this study is the first to elucidate the causal relationship between plasma amyloid marker levels and CSVD traits from a genetic perspective. This study enhances our understanding of the pathophysiologic mechanisms of CSVD, especially amyloid deposition, and offers potential targets for treatment. The findings of this study hold multifaceted potential clinical applications. First, plasma amyloid biomarker levels, particularly plasma Aβ, can serve as risk stratification indicators to help identify individuals with higher genetic risk. Second, they may be utilized in future clinical trials for screening or stratified enrollment to evaluate the efficacy of anti-amyloid or related pathway drugs in CSVD. Finally, the combined dynamic monitoring of plasma amyloid biomarkers and MRI imaging may facilitate the early detection of microbleed progression, thereby providing a basis for personalized interventions.

This study has several limitations. The p value cut-off for SNP selection was set at 1 × 10⁻⁵, which may have resulted in weak IV bias. This may have affected the results. Furthermore, our results are mainly based on data from a European population and therefore may not be generalizable to other populations. In addition, the ICH GWAS database we used was not separated into lobar and nonlobar hemorrhages according to the site of hemorrhage, so we could not analyze the relationship between Aβ levels and the risk of developing cerebral hemorrhage at different sites (implying different etiologies). Finally, GWASs using ICH, SVS, CMBs, and WMH genetic variants may not fully capture the overall status of CSVD. Since the GWAS exposure identified only one eligible instrumental variable under strict significance thresholds (p < 5 × 10⁻⁸), the MR estimates in this study essentially correspond to Wald ratio estimates. Although the results demonstrate statistical significance, their causal inferential power is limited due to the inability to test for pleiotropy and heterogeneity. These findings should be interpreted with caution, and future validation using larger samples or more lenient thresholds is recommended.

In conclusion, our study provides suggestive evidence that plasma amyloid marker levels may increase the risk of developing CSVD. Future RCTs are necessary to validate these findings. Our results contribute to a greater understanding of the pathophysiology of CSVD, particularly in relation to amyloid deposition, and offer insights into potential biomarkers and therapeutic targets.

Supplemental Material

sj-docx-1-alr-10.1177_25424823251403565 - Supplemental material for Amyloid marker levels and the risk of developing cerebral small vessel disease: A Mendelian randomization study

Supplemental material, sj-docx-1-alr-10.1177_25424823251403565 for Amyloid marker levels and the risk of developing cerebral small vessel disease: A Mendelian randomization study by Zhuoya Wang, Kailin Xia, Xin Huang, Dongsheng Fan and Qiong Yang in Journal of Alzheimer's Disease Reports

Footnotes

Acknowledgements

We thank all GWAS studies for publicly providing summary data and express our gratitude to all participants who contributed to these studies. We also appreciate the support from the National Natural Science Foundation of China.

ORCID iDs

Qiong Yang

Zhuoya Wang

Ethical considerations

The Mendelian randomization approach was adopted in this study. All research data were derived from publicly published datasets, which had obtained ethical approval.

Consent to participate

This study was based on publicly available summary-level GWAS data, and no individual-level data were used. Therefore, additional informed consent from participants was not required.

Consent for publication

Not applicable

Author contribution(s)

Zhuoya Wang: Formal analysis; Investigation; Methodology; Resources; Software; Writing – original draft.

Kailin Xia: Methodology; Supervision; Validation.

Xin Huang: Validation.

Dongsheng Fan: Conceptualization; Formal analysis; Writing – review & editing.

Qiong Yang: Conceptualization; Formal analysis; Writing – review & editing.

Funding

QY was funded by the National Natural Science Foundation of China (81901204), Clinical Cohort Construction Program of Peking University Third Hospital (BYSYDL2024019). DF was funded by the National Natural Science Foundation of China (81873784 and 82071426), Clinical Cohort Construction Program of Peking University Third Hospital (BYSYDL2019002 and BYSYZD2021004). This work was supported by the special fund of the National Clinical Key Specialty Construction Program, P. R. China (2024).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The exposure and outcome GWASs are available in the corresponding previous research.

Supplemental material

Supplemental material for this article is available online.

References

Wardlaw

Smith

Dichgans

. Small vessel disease: mechanisms and clinical implications. Lancet Neurol 2019; 18: 684–696.

Ter Telgte

van Leijsen

EMC

Wiegertjes

, et al. Cerebral small vessel disease: from a focal to a global perspective. Nat Rev Neurol 2018; 14: 387–398.

Wardlaw

Smith

Biessels

, et al. Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lancet Neurol 2013; 12: 822–838.

Markus

de Leeuw

. Cerebral small vessel disease: recent advances and future directions. Int J Stroke 2023; 18: 4–14.

Vernooij

van der Lugt

Ikram

, et al. Prevalence and risk factors of cerebral microbleeds: the Rotterdam Scan Study. Neurology 2008; 70: 1208–1214.

Woo

Sauerbeck

Kissela

, et al. Genetic and environmental risk factors for intracerebral hemorrhage: preliminary results of a population-based study. Stroke 2002; 33: 1190–1195.

Greenberg

Vernooij

Cordonnier

, et al. Cerebral microbleeds: a guide to detection and interpretation. Lancet Neurol 2009; 8: 165–174.

Charidimou

Boulouis

Gurol

, et al. Emerging concepts in sporadic cerebral amyloid angiopathy. Brain 2017; 140: 1829–1850.

Koch

Enzlein

Chen

, et al. APP Substrate ectodomain defines amyloid-β peptide length by restraining γ-secretase processivity and facilitating product release. EMBO J 2023; 42: e114372.

10.

Mold

Shrive

Exley

. Serum amyloid P component accelerates the formation and enhances the stability of amyloid fibrils in a physiologically significant under-saturated solution of amyloid-β42. J Alzheimers Dis 2012; 29: 875–881.

11.

Greenberg

Bacskai

Hernandez-Guillamon

, et al. Cerebral amyloid angiopathy and Alzheimer disease - one peptide, two pathways. Nat Rev Neurol 2020; 16: 30–42.

12.

Tennent

Lovat

Pepys

. Serum amyloid P component prevents proteolysis of the amyloid fibrils of Alzheimer disease and systemic amyloidosis. Proc Natl Acad Sci U S A 1995; 92: 4299–4303.

13.

Knopman

Amieva

Petersen

, et al. Alzheimer disease. Nat Rev Dis Primers 2021; 7: 33.

14.

Zetterberg

. Biofluid-based biomarkers for Alzheimer's disease-related pathologies: an update and synthesis of the literature. Alzheimers Dement 2022; 18: 1687–1693.

15.

Gurol

Irizarry

Smith

, et al. Plasma beta-amyloid and white matter lesions in AD, MCI, and cerebral amyloid angiopathy. Neurology 2006; 66: 23–29.

16.

Janelidze

Stomrud

Palmqvist

, et al. Plasma β-amyloid in Alzheimer's disease and vascular disease. Sci Rep 2016; 6: 26801.

17.

van Dijk

Prins

Vermeer

, et al. Plasma amyloid beta, apolipoprotein E, lacunar infarcts, and white matter lesions. Ann Neurol 2004; 55: 570–575.

18.

Toledo

Vanderstichele

Figurski

, et al. Factors affecting Aβ plasma levels and their utility as biomarkers in ADNI. Acta Neuropathol 2011; 122: 401–413.

19.

van Leijsen

EMC

Kuiperij

Kersten

, et al. Plasma Aβ (amyloid-β) levels and severity and progression of small vessel disease. Stroke 2018; 49: 884–890.

20.

Kim

Park

Yun

, et al. Screening for cerebral amyloid angiopathy based on serological biomarkers analysis using a dielectrophoretic force-driven biosensor platform. Lab Chip 2021; 21: 4557–4565.

21.

Piccarducci

Caselli

Zappelli

, et al. The role of amyloid-β, tau, and α-synuclein proteins as putative blood biomarkers in patients with cerebral amyloid angiopathy. J Alzheimers Dis 2022; 89: 1039–1049.

22.

Greenberg

Cho

O'Donnell

, et al. Plasma beta-amyloid peptide, transforming growth factor-beta 1, and risk for cerebral amyloid angiopathy. Ann N Y Acad Sci 2000; 903: 144–149.

23.

Smith

Ebrahim

. ‘Mendelian randomization’: can genetic epidemiology contribute to understanding environmental determinants of disease? Int J Epidemiol 2003; 32: 1–22.

24.

Haycock

Burgess

Wade

, et al. Best (but oft-forgotten) practices: the design, analysis, and interpretation of Mendelian randomization studies. Am J Clin Nutr 2016; 103: 965–978.

25.

Chai

Cheah

, et al. Blood-based biomarkers of cerebral small vessel disease. Ageing Res Rev 2024; 95: 102247.

26.

Sun

Maranville

Peters

, et al. Genomic atlas of the human plasma proteome. Nature 2018; 558: 73–79.

27.

Ferkingstad

Sulem

Atlason

, et al. Large-scale integration of the plasma proteome with genetics and disease. Nat Genet 2021; 53: 1712–1721.

28.

Chouraki

De Bruijn

Chapuis

, et al. A genome-wide association meta-analysis of plasma Aβ peptides concentrations in the elderly. Mol Psychiatry 2014; 19: 1326–1335.

29.

Larsson

Chen

Gill

, et al. Risk factors for intracerebral hemorrhage: genome-wide association study and Mendelian randomization analyses. Stroke 2024; 55: 1582–1591.

30.

Mishra

Malik

Hachiya

, et al. Stroke genetics informs drug discovery and risk prediction across ancestries. Nature 2022; 611: 115–123.

31.

Knol

Traylor

, et al. Association of common genetic variants with brain microbleeds: a genome-wide association study. Neurology 2020; 95: e3331–e3343.

32.

Zhao

Yang

, et al. Common genetic variation influencing human white matter microstructure. Science 2021; 372: eabf3736.

33.

Pierce

Ahsan

Vanderweele

. Power and instrument strength requirements for Mendelian randomization studies using multiple genetic variants. Int J Epidemiol 2011; 40: 740–752.

34.

Bowden

Del Greco

Minelli

, et al. Assessing the suitability of summary data for two-sample Mendelian randomization analyses using MR-Egger regression: the role of the I2 statistic. Int J Epidemiol 2016; 45: 1961–1974.

35.

Pierce

Burgess

. Efficient design for Mendelian randomization studies: subsample and 2-sample instrumental variable estimators. Am J Epidemiol 2013; 178: 1177–1184.

36.

Zhang

, et al. Lipids, apolipoproteins, statins, and intracerebral hemorrhage: a Mendelian randomization study. Ann Neurol 2022; 92: 390–399.

37.

Xia

Wang

, et al. Essential nutrients and cerebral small vessel diseases: a two-sample Mendelian randomization study. Front Nutr 2023; 10: 1172587.

38.

Wang

Xia

, et al. Essential nutrients and white matter hyperintensities: a two-sample Mendelian randomization study. Biomedicines 2024; 12: 810.

39.

Burgess

Bowden

Fall

, et al. Sensitivity analyses for robust causal inference from Mendelian randomization analyses with multiple genetic variants. Epidemiology 2017; 28: 30–42.

40.

Bowden

Davey Smith

Burgess

. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol 2015; 44: 512–525.

41.

Bowden

Davey Smith

Haycock

, et al. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol 2016; 40: 304–314.

42.

Sanderson

Spiller

Bowden

. Testing and correcting for weak and pleiotropic instruments in two-sample multivariable Mendelian randomization. Stat Med 2021; 40: 5434–5452.

43.

Jung

Jang

Park

, et al. Strictly lobar microbleeds reflect amyloid angiopathy regardless of cerebral and cerebellar compartments. Stroke 2020; 51: 3600–3607.

44.

Hernandez-Guillamon

Delgado

Penalba

, et al. Plasma β-amyloid levels in cerebral amyloid angiopathy-associated hemorrhagic stroke. Neurodegener Dis 2012; 10: 320–323.

45.

Chatterjee

Tegg

Pedrini

, et al. Plasma amyloid-beta levels in a pre-symptomatic Dutch-type hereditary cerebral amyloid angiopathy pedigree: a cross-sectional and longitudinal investigation. Int J Mol Sci 2021; 22: 2931.

46.

Calhoun

Burgermeister

Phinney

, et al. Neuronal overexpression of mutant amyloid precursor protein results in prominent deposition of cerebrovascular amyloid. Proc Natl Acad Sci U S A 1999; 96: 14088–14093.

47.

Koemans

Chhatwal

van Veluw

, et al. Progression of cerebral amyloid angiopathy: a pathophysiological framework. Lancet Neurol 2023; 22: 632–642.

48.

Cordonnier

van der Flier

. Brain microbleeds and Alzheimer's disease: innocent observation or key player? Brain 2011; 134: 335–344.

49.

Jarrett

Berger

Lansbury

Jr.

The carboxy terminus of the beta amyloid protein is critical for the seeding of amyloid formation: implications for the pathogenesis of Alzheimer's disease. Biochemistry 1993; 32: 4693–4697.

50.

Hawrylycz

Lein

Guillozet-Bongaarts

, et al. An anatomically comprehensive atlas of the adult human brain transcriptome. Nature 2012; 489: 391–399.

51.

Hondius

Eigenhuis

Morrema

THJ

, et al. Proteomics analysis identifies new markers associated with capillary cerebral amyloid angiopathy in Alzheimer's disease. Acta Neuropathol Commun 2018; 6: 46.

52.

Nagatoshi

Ueda

, et al. Serum amyloid P component: a novel potential player in vessel degeneration in CADASIL. J Neurol Sci 2017; 379: 69–76.

53.

Cupino

Zabel

. Alzheimer's silent partner: cerebral amyloid angiopathy. Transl Stroke Res 2014; 5: 330–337.

54.

Ellmerich

Taylor

Richardson

, et al. Dementia in the older population is associated with neocortex content of serum amyloid P component. Brain Commun 2021; 3: fcab225.

55.

Schmidt

Finan

Chopade

, et al. Genetic evidence for serum amyloid P component as a drug target in neurodegenerative disorders. Open Biol 2024; 14: 230419.

56.

O'Brien

Wong

. Amyloid precursor protein processing and Alzheimer's disease. Annu Rev Neurosci 2011; 34: 185–204.

57.

Klakotskaia

Agca

Richardson

, et al. Memory deficiency, cerebral amyloid angiopathy, and amyloid-β plaques in APP+PS1 double transgenic rat model of Alzheimer's disease. PLoS One 2018; 13: e0195469.

58.

Xiaobin

Hui

, et al. An APP mutation family exhibiting white matter hyperintensities and cortical calcification in East China. Neurol Sci 2020; 41: 2921–2928.

59.

Padovani

Pastorino

Borroni

, et al. Amyloid precursor protein in platelets: a peripheral marker for the diagnosis of sporadic AD. Neurology 2001; 57: 2243–2248.

60.

Hampel

Elhage

Cho

, et al. Amyloid-related imaging abnormalities (ARIA): radiological, biological and clinical characteristics. Brain 2023; 146: 4414–4424.

61.

Wheeler

Irimia

Braskie

. Using neuroimaging to study cerebral amyloid angiopathy and its relationship to Alzheimer's disease. J Alzheimers Dis 2024; 97: 1479–1502.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

45.97 MB