Preclinical amyloid‐PET SUVr in the precuneus and posterior cingulate associations with MoCA scores are driven by language domain subscores

Abstract

AbstractBackgroundβ‐amyloid (Aβ) deposition in the brain is the earliest pathological change supporting a diagnosis of preclinical Alzheimer’s disease (pAD). Such deposition can be quantified using amyloid‐PET scans, which also enable a quantitative evaluation of the distribution of the pathology in pAD. Cognitive screening instruments such as the Montreal Cognitive Assessment (MoCA) are widely used to detect cognitive decline as one transitions from pAD to mild cognitive impairment or dementia due to AD, but have little proven utility in tracking cognitive decline during the stage of pAD. The present study sought to identify potential relationships between regional Aβ deposition and domain specific subscales and items comprising the MoCA in pAD.MethodNinety‐nine clinically normal participants with a MoCA score ≥ 23 underwent F18‐florbetapir PET/CT scanning and regional standardized uptake value ratio (SUVr) was determined using the Siemens Syngo.via® Neurology software package. Relationships between regional SUVr and MoCA test performance as well as MoCA domain scores were analyzed using linear regression models adjusted for age and education.ResultSubjects included 68 women and 31 men with a mean age of 74.2±6.3 years, education 16.7±2.6 years, and MoCA 26.6± 1.9. The adjusted linear regression analyses demonstrated that reduced global MoCA score was associated with increased regional SUVr in the precuneus and posterior cingulate regions only (p<0.05). This relationship appeared to be driven predominantly by the language domain index score rather than other subdomains of the MoCA (R2=0.122; p= 0.003).ConclusionThese data demonstrate that increased Aβ deposition in the precuneus and posterior cingulate (the earliest brain regions affected with Aβ pathology) are associated with decreased global cognitive performance, driven by language domain subscores in individuals with pAD. Further research using more sensitive domain‐specific tests in cognitively normal individuals may help determine optimal low‐cost, non‐invasive, neuropsychological predictors of pAD.