Exploring biomarker profiles of the cognitively unimpaired individuals in early tau PET Braak stage.

Background: In this study, we evaluated the association between tau PET and fluid biomarkers and explored biological mechanisms associated with elevated tau in cognitively unimpaired(CU) individuals. Method: Seventy‐nine CU individuals from the ALFA+ cohort had [18F]RO‐948 and [18F]flutemetamol PET, T1‐weighted MRI, CSF and plasma biomarkers. Longitudinal changes in cognition were measured using cognition composites(Table1). Participants were categorized into AT stages using CSF biomarkers and pre‐established cut‐offs(Table 1). [18F]RO‐948 SUVR was measured in entorhinal(BraakI/II), limbic(BraakIII/IV), and neocortical(BraakV/VI) regions using the inferior cerebellum as reference. Regional positivity thresholds per Braak region were calculated as mean+2SD of the A‐T‐ group. [18F]flutemetamol PET scans were quantified in Centiloids(CL). Associations between [18F]RO‐948 SUVRs and AD biomarkers (Figure1) were assessed using Pearson correlations. Receiver Operating Curve(ROC) analyses were conducted to determine the capacity of biomarkers to predict BraakI/II positivity. In those with CL>cut‐off for tau‐PET positivity, differences in non‐core AD fluid biomarkers(Figure2) and longitudinal cognition were sought between tau BraakI/II positive and negative individuals. The mediating effect of non‐core AD biomarkers on the AD pathophysiological cascade were sought. FDR‐corrected p‐values<0.05 were considered significant. Result: Nine cases(11.4%) were positive in BraakI/II. Following a hierarchical pattern, four of them were also positive for BraakIII/IV and one for BraakV/VI. Fluid biomarkers presented correlations with tau PET SUVR in all Braak stages (Figure1) and predictive capacity for tau‐PET positivity in BraakI/II(Table2). In individuals with CL>32.53, BraakI/II positives had higher levels of CSF‐GFAP(p<0.01) and CSF‐NFL(p = 0.03), and presented declines in PACC and in visual, attention, and memory composites (Figure1). CSF‐GFAP partially mediated the association between CL and tau‐PET (16%) and between tau‐PET and CSF‐NfL (30%). Conclusion: Early in the AD continuum, [18F]RO‐948 PET conformed to the Braak hierarchical model. CSF and plasma biomarkers showed moderate associations with tau PET SUVR but good capacity to predict tau PET positivity in BraakI/II. Tau‐PET positivity in this region was associated with higher astrogliosis, neurodegeneration and cognitive decline as compared to tau‐PET negatives with similar levels of amyloid deposition. Astrogliosis partially explained the observed associations between amyloid deposition and the presence of tau tangles in medial temporal regions and, even strongly, between the latter and neurodegeneration. [ABSTRACT FROM AUTHOR]