Amyloid‐β CSF/PET discordance vs tau load 5 years later: It takes two to tangle

Abstract

AbstractBackgroundAmyloid‐β pathology assessed in vivo using CSF and PET results in discordance (CSF+/PET‐ or CSF‐/PET+) in 10‐20% of cases. An amyloid‐β CSF+/PET‐ profile has been proposed to mark the pathological beginnings of amyloid‐β accumulation (Palmqvist et al., 2016), providing a powerful model to study the interplay between amyloid‐β and tau at the earliest stages of Alzheimer’s disease.MethodsWe included 730 participants without dementia (282 cognitively normal, 448 MCI) from the Alzheimer’s Disease Neuroimaging Initiative database with available baseline [18F]florbetapir PET and CSF Aβ42 within one year. Established cut‐offs were used to determine amyloid‐β CSF/PET status. We used linear mixed modelling to study the association between amyloid‐β CSF/PET status and tau pathology measured in CSF and cognitive performance (both available at baseline) or by using [18F]flortaucipir tau‐PET (performed 5.2±1.2 years later). Using Kaplan‐Meier estimates and Cox regression, we investigated the association between amyloid‐β CSF/PET status and progression of syndromic diagnosis. Additionally, we calculated the proportion of CSF+/PET‐ participants who during follow‐up (i) progressed to amyloid‐β CSF+/PET+, and/or (ii) became tau‐positive based on [18F]flortaucipir PET.ResultsMean age of the participants was 73±7 years, 52% were male and mean MMSE was 28.4±1.6. (Table 1). Amyloid‐β CSF+/PET+ (N=318) participants had elevated CSF (p)tau levels and worse cognitive performance at baseline, while CSF+/PET‐ (N=80) participants were overall similar to the CSF‐/PET‐ (N=306) group (Figure 1). Five years after baseline, [18F]flortaucipir PET uptake in the CSF+/PET‐ group (1.20±0.13) did not differ from CSF‐/PET‐ (1.18±0.08, p=0.69), but was substantially lower than CSF+/PET+ (1.48±0.44, p<0.001). Progression of syndromic diagnosis occurred less often in amyloid‐β CSF+/PET‐ (14%, hazard ratio [HR]=0.22[0.12,0.41], p<0.001), CSF‐/PET+ (8%, HR=0.10[0.02,0.41], p=0.001) and CSF‐/PET‐ (10%, HR=0.16[0.11,0.24], p<0.001) participants, compared to the CSF+/PET+ (44%) group (Figure 2). Of the CSF+/PET‐ subjects, 21/64 (33%) progressed to amyloid‐β CSF+/PET+, whereas only one (3%, difference p<0.05) became tau‐positive based on [18F]flortaucipir PET.ConclusionsSufficient amyloid‐β load detectable by both CSF and PET is required before substantial tau deposition emerges. Compared to participants with abnormal amyloid‐β levels on PET and CSF, the discordant CSF/PET groups have a distinctly better prognosis.