Background: We evaluated the utility of longitudinal measures of plasma amyloid-β (Aβ) as a means to identify pre-symptomatic cognitive decline in Alzheimer's disease (AD) when coupled to neuroimaging and neuropsychological parameters.
Methods: Participants from the Australian Imaging, Biomarkers and Lifestyle (AIBL) study were grouped based upon cognitive change and changes in measurable levels of neocortical amyloid over 36 months. Participants were classified as those who transitioned for cognitive decline and change in neocortical amyloid, those healthy controls that did not transition, and stable AD participants over 36 months.
Results: Comparisons of plasma Aβ levels between the transition and non-transitional groups showed Aβ1-42 and the Aβ1-42/Aβ1-40 ratio were significantly decreased at baseline (p = 0.008 and p = 0.002, respectively) and at 18 months (p = 0.003 and p = 0.004, respectively). Both measures of neocortical amyloid and two previously published composite scores validated the creation of the novel transitional grouping (p < 0.0001). In addition Aβn-42 performed well as a longitudinal prognostic indicator of transition toward cognitive decline, with a significant decrease in the transition group at the 18 month time point (p = 0.01).
Conclusion: We demonstrated that baseline plasma Aβ1-42 and the Aβ1-42/Aβ1-40 ratio were putative biomarkers indicative of cognitive decline and validated this result using 18 month data. We created a novel transitional grouping and validated this measure using published measures of neocortical amyloid and composite memory scores. These findings suggest that longitudinal plasma Aβ could contribute to a pre-symptomatic biomarker panel for AD.
Keywords: Alzheimer's disease; Pittsburgh compound B; biomarkers; diagnosis; neocortical amyloid burden; plasma amyloid-β; positron emission topography.