Fig. 5
Baseline plasma NTA-tau association with longitudinal tau-PET and neurodegeneration. Associations between baseline plasma NTA-tau levels and longitudinal tau-PET (A) and cortical thickness determined through MRI (B and C, BioFINDER-2 and -1 respectively). We used linear mixed models with tau-PET (SUVR) and cortical thickness (mm) as outcome and the interaction of baseline plasma biomarkers and time as predictor with random intercepts and random time-slopes. Age and sex were used as covariates. Dots and thin lines represent individual timepoints and trajectories, respectively, for each participant. Each participant is coloured based on its baseline plasma NTA-tau levels. Thick lines and shaded areas represent the mean trajectory over time of each group of plasma NTA-tau baseline levels and its 95%CI. Standardized β (βstd) and p-values of the associations as well as the R2 of the model are shown in the plots. Only Aβ+ within the AD continuum (excluding nonAD+) were included in these analyses, as were those expected to progress. Standardized β (βstd) and p-values of the associations as well as the R2 of the model are shown in the plots. *: p < 0.05; **: p < 0.01; ***: p < 0.001. Abbreviations: Aβ, amyloid-β; AD, Alzheimer’s disease; CI, confidence interval; nonAD+ ; non-Alzheimer’s type dementia Aβ-positive; SUVR, standardized uptake value ratio