Amyloid‐independent vascular contributions to cortical atrophy and cognition in a multi‐center mixed cohort with low to severe small vessel disease
Journal Articles
Overview
Research
Identity
Additional Document Info
View All
Overview
abstract
AbstractBackgroundSmall vessel disease (SVD) often co‐exists with Alzheimer’s disease (AD) pathology (up to 60%) and may facilitate AD progression. However, SVD is currently not integrated as a pathological factor within the ATN research criteria. Up to now, the majority of studies that investigated effects of SVD on brain atrophy and cognition were limited to either AD cohorts with low SVD burden (e.g., ADNI) or cognitively normal elderly with high SVD burden. Thus, there is a need to investigate the effects of SVD in a cohort spanning low to severe SVD and amyloid‐beta pathology.MethodOur study included 118 subjects in total. Fifty‐nine subjects were recruited in a multi‐site study (MITNEC) from dementia and SVD‐stroke clinics (64% amyloid‐beta+) who had severe SVD burden as quantified by white matter hyperintensity volumes [WMH; median(IQR): 30.2(22.2)cm3] and Fazekas score 2.5‐3. In addition, we included 59 cognitively normal/early‐MCI subjects from ADNI (44% amyloid‐beta+) with low‐to‐moderate WMH [median(IQR): 5.8(9.1)cm3]. We performed vertex‐wise regressions, investigating associations of cortical thickness with amyloid‐beta (18F‐florbetapir‐SUVRpons) or vascular burden (WMH), corrected for WMH or amyloid‐beta, respectively, and age, sex, education. Further, mediation analyses investigated whether the effects of amyloid‐beta or vascular burden on cognition (MMSE, MoCA, Trails‐B, semantic fluency, ANART, and Boston‐naming) were mediated by cortical thickness.ResultWe observed a significant effect of vascular burden on cortical thickness independent of amyloid‐beta. This effect was stronger than the vascular‐independent effect of amyloid‐beta on thickness (Fig. 1). Furthermore, we observed additive effects of vascular burden and amyloid‐beta on cognition (semantic fluency: ß=‐0.29 [p=0.002] and ß=‐0.27 [p=0.001], respectively; Trails‐B: ß=+0.18 [p=0.03] and ß=+0.22 [p=0.02], respectively). Here, the effects on semantic fluency were significantly mediated by cortical thickness (vascular→thickness→semantic: 35% mediation, 95%CI[‐0.19,‐0.02]; amyloid‐beta→thickness→semantic: 19% mediation, 95%CI[‐0.11,‐0.003]) (Fig. 1).ConclusionIn our study of mixed AD/SVD and control subjects, the effect of SVD burden (WMH) exceeded the effect of amyloid‐beta on neurodegeneration alone. Furthermore, vascular burden contributed to semantic loss both directly and through its impact on neurodegeneration. As such, the presence of cerebrovascular comorbidities supports the idea of combinational therapeutic approaches where SVD factors may be targeted alongside amyloid‐beta to halt neurodegeneration and cognitive decline.
