In vivo amyloid progression in healthy middle‐aged to older people at risk of Alzheimer’s disease

Abstract

AbstractBackgroundPrevious studies applied in‐vivo amyloid staging to amyloid‐sensitive PET scans of healthy older people as well as patients with Alzheimer’s disease (AD) dementia or mild cognitive impairment to characterize stages of amyloid deposition. We adapted this approach to a sample of healthy middle‐aged to older participants at elevated risk for AD due to parental history of AD dementia.MethodWe analyzed 11C‐PiB‐PET scans of 220 participants enrolled in the Wisconsin Registry for Alzheimer’s Prevention (WRAP) (mean age = 61.4y, range: 46‐76y). Distribution volume ratio (DVR) maps were calculated from the dynamic PIB‐PET acquisitions and corrected for partial volume effects. DVR values were extracted from 52 cortical and subcortical regions defined in the Harvard‐Oxford atlas and were used together with global mean DVR values in two‐dimensional Gaussian Mixture Modeling to establish regional thresholds for amyloid‐positivity. Resampled regional frequencies of amyloid‐positivity across participants were used to estimate a four‐stage model of regional amyloid accumulation. We staged individual DVR maps, and assessed the associations between amyloid stage and episodic memory (delayed wordlist recall) and executive function performance (TMT‐B) with Spearman correlation.ResultAccording to the amyloid progression model, stage I involved the anterior and posterior cingulate as well as lateral temporal cortex; stage II had more extensive frontal and temporo‐parietal involvement; stage III proceeded into occipito‐parietal cortical regions and the hippocampus; stage IV extended to the occipital pole, pre‐ and postcentral gyri, and subcortical structures (Figure 1). Out of 220 participants, 36 (16%) were assigned to one of the four amyloid stages, and two cases (1%) deviated from the hierarchical pattern of amyloid accumulation. Higher amyloid stages (I‐IV) correlated with lower executive function (rho = 0.35, p = 0.043), but not with lower episodic memory performance (rho = 0.13, p = 0.48).ConclusionThe estimated 11C‐PiB‐PET‐based amyloid staging model is largely consistent with recently reported amyloid staging models for [F18]‐based amyloid‐PET scans. In this relatively young sample of healthy participants at elevated risk of AD, advanced amyloid stages were rare and not consistently linked to decreased cognitive performance. In future work we will assess whether the derived amyloid stages translate into a stage‐proportional risk of longitudinal cognitive decline in this population.