The association between white matter hyperintensities and amyloid and tau deposition

Abstract

White matter hyperintensities (WMHs) frequently occur in Alzheimer's Disease (AD) and have a contribution from ischemia, though their relationship with β-amyloid and cardiovascular risk factors (CVRFs) is not completely understood. We used AT classification to categorize individuals based on their β-amyloid and tau pathologies, then assessed the effects of β-amyloid and tau on WMH volume and number. We then determined regions in which β-amyloid and WMH accumulation were related. Last, we analyzed the effects of various CVRFs on WMHs. As secondary analyses, we observed effects of age and sex differences, atrophy, cognitive scores, and APOE genotype. PET, MRI, FLAIR, demographic, and cardiovascular health data was collected from the Alzheimer's Disease Neuroimaging Initiative (ADNI-3) (N = 287, 48 % male). Participants were categorized as A + and T + if their Florbetapir SUVR and Flortaucipir SUVR were above 0.79 and 1.25, respectively. WMHs were mapped on MRI using a deep convolutional neural network (Sepehrband et al., 2020). CVRF scores were based on history of hypertension, systolic and diastolic blood pressure, pulse rate, respiration rate, BMI, and a cumulative score with 6 being the maximum score. Regression models and Pearson correlations were used to test associations and correlations between variables, respectively, with age, sex, years of education, and scanner manufacturer as covariates of no interest. WMH volume percent was significantly associated with global β-amyloid (r = 0.28, p < 0.001), but not tau (r = 0.05, p = 0.25). WMH volume percent was higher in individuals with either A + or T + pathology compared to controls, particularly within in the A+/T + group (p = 0.007, Cohen's d = 0.4, t = -2.5). Individual CVRFs nor cumulative CVRF scores were associated with increased WMH volume. Finally, the regions where β-amyloid and WMH count were most positively associated were the middle temporal region in the right hemisphere (r = 0.18, p = 0.002) and the fusiform region in the left hemisphere (r = 0.017, p = 0.005). β-amyloid and WMH have a clear association, though the mechanism facilitating this association is still not fully understood. The associations found between β-amyloid and WMH burden emphasizes the relationship between β-amyloid and vascular lesion formation while factors like CVRFs, age, and sex affect AD development through various mechanisms. These findings highlight potential causes and mechanisms of AD as targets for future preventions and treatments. Going forward, a larger emphasis may be placed on β-amyloid's vascular effects and the implications of impaired brain clearance in AD.

Fig. 1

Correlation between white matter hyperintensity volume and global β-amyloid standard uptake value ratio in all participants (r = 0.28, p < 0.001).

Fig. 2

Mean WMH volume differences between Amyloid negative, Tau negative (A-/T-) participants and participants with different pathology status: Amyloid positive, Tau negative (A+/T-; (p = 0.002, Cohen’s d = 0.5); Amyloid negative, Tau positive (A-/T+; (p = 0.02, Cohen’s d = 0.9); Amyloid positive, Tau positive (A+/T+; p = 0.007, Cohen’s d = 0.4).

Fig. 3

A. Correlation between White matter hyperintensities (WMH) volume correlation and global Amyloid uptake among different pathology groups. Significant positive correlation in A+/T- (r = 0.36, p < 0.001), marginal positive correlation in A-/T+ (r = 0.58, p = 0.0498), nonsignificant correlation in A+/T+ (r = 0.23, p = 0.09), and no significant correlation in A-/T- (r = 0.06, p = 0.45). B. Correlation between WMH count and global Amyloid uptake among different pathology groups. Nonsignificant positive correlations in A+/T- (r = 0.14, p = 0.22), A-/T+ (r = 0.35, p = 0.27), and A-/T- (r = 0.17, p = 0.07). A significant positive association in A+/T+ (r = 0.33, p = 0.01).

Fig. 4

Correlation between white matter hyperintensities (WMH) volume and meta-temporal Tau standard uptake value ratio (correlation p = 0.25; association p = 0.4).

Fig. 5

A. Correlation between white matter hyperintensity (WMH) count and global Amyloid standard uptake value ratio (p < 0.001, r = 0.25). B. Correlation between white matter hyperintensity (WMH) count and meta-temporal Tau standard uptake value ratio (p = 0.02, r = 0.11).

Fig. 6

Mean WMH count differences between groups. Amyloid positive, tau positive (A+/T+; p = 0.001, Cohen’s d = 0.5); Amyloid positive, tau negative (A+/T-; p = 0.007, Cohen’s d = 0.4); Amyloid negative, tau positive (A-/T+; p = 0.04, Cohen’s d = 0.6).

Fig. 7

Regional Pearson correlation between β-amyloid uptake and white matter hyperintensities in left and right hemispheres, showing p-values, R-values, and only significant R-values (p < 0.05). The most statistically significant regions are indicated by red arrows. Right hemisphere: middle temporal (p = 0.002, R = 0.18); Left hemisphere: fusiform (p = 0.005, R = 0.017). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)