White matter integrity disruption in early amyloid accumulators

Abstract

AbstractBackgroundAmyloid‐β (Aβ) accumulation is the pathological hallmark of Alzheimer’s disease (AD), suggested to be closely followed by white matter (WM) integrity alterations in preclinical subjects. However, reports on the relationship between the two measures have been conflicting. This study aimed to investigate the relationship between amyloid burden and WM integrity in preclinical AD on a regional level.Methods179 subjects from the EMIF‐AD preclinAD study who underwent Diffusion Weighted MRI (DWI) and dynamic [18F]flutemetamol PET imaging (Philips Ingenuity PET‐MR) were included (Table 1). Global and tract‐level Fractional anisotropy (FA) and mean diffusivity (MD) were determined using the JHU WM atlas. Amyloid burden was determined by the non‐displaceable binding potential (BPND+1) with cerebellar grey matter as the reference region. The relationship between amyloid burden and WM integrity was tested using Generalized estimating equations, correcting for family relatedness (Figure 1). Linear and quadratic fits were tested using QIC. Analyses were corrected for age, sex, and white matter hyperintensity load. The strongest associations are shown (significance set at p<0.05, Bonferroni‐corrected).ResultsLow amyloid burden was associated with increased FA and decreased MD, followed by decreased FA and increased MD upon higher amyloid burden. This relationship was most robustly observed in the body of the corpus callosum (body CC), for which a quadratic fit was preferred according to QIC, and was mainly driven by amyloid accumulation in the precuneus (FA: β linear=0.352, β quadratic=‐0.103, ΔQIC=6.7; MD: β linear=‐0.434, β quadratic=0.128, ΔQIC=8.0, Figure 2), PCC (MD: β linear=‐0.393, β quadratic=0.101, ΔQIC=4.0), and to a lesser extent by the insula (MD: β linear=‐0.256, β quadratic=0.063, ΔQIC=2.0).ConclusionEarly amyloid deposition is associated with WM changes, particularly evident in the body CC. Importantly, a quadratic fit was preferred by QIC, and this pattern might reflect multiple stages of axonal damage, with initial oligodendrocyte proliferation and axonal swelling upon lower amyloid burden, and actual axonal degeneration only becoming apparent at a higher amyloid burden. Thus, WM integrity specifically in the body CC could be considered as an indicator of early AD pathology in addition to amyloid burden in preclinical subjects.