Estimating the time between Aβ positivity and elevated regional tau in preclinical AD

Abstract Background Understanding when Aß positive cognitively normal individuals develop tau pathology has important implications for treatment with anti‐Aß therapies. We employed a changepoint regression approach to estimate time from Aß‐PET positivity to regionally elevated tau‐PET in a population‐based cohort of primarily cognitively unimpaired individuals. Method Participants from Insight 46 (1946 British birth cohort) underwent two [18F]florbetapir Aß‐PET scans and a sub‐sample enriched for Aß were also scanned with [18F]MK‐6240 tau‐PET, characteristics are presented in Table 1. Age at Aß‐PET positivity, defined through Gaussian mixture modelling (GMM) as >12 Centiloids, was estimated using the SILA algorithm (Betthauser et al. 2022). Tau‐PET SUVRs from a temporal‐meta, Braak I‐VI and 34 bilateral FreeSurfer7 regions were calculated 90‐110 mins post‐injection using an inferior cerebellar grey matter reference region. Tau‐PET positivity cutpoints for each region were defined using the 99th percentile of the lower distribution from GMM. For each region, we tested changepoints in the slope of the linear regression between tau‐PET SUVR and time from estimated Aß positivity with Bonferroni correction (5% significance,p<.001). Result At tau‐PET scan, 90 individuals (60%) were >0 years from estimated Aß positivity. There was a statistically significant difference in slopes before and after the changepoint for all composite regions excluding Braak VI (Figure 1). Changepoint estimates show that tau accumulation started ∼2.5yrs after estimated Aß positivity in Braak stages I‐III; and ∼10‐12.5yrs for Braak stages IV‐V, and the temporal‐meta region (Figure 2A). Regional estimates broadly followed Braak stages for 16/36 regions showing significant evidence of change in slopes (Figure 2B). Tau positivity rates before changepoints did not exceed 5%, whereas post‐changepoint tau positivity rates were between ∼10% and ∼50%. Regions with later changepoints tended to exhibit higher post‐changepoint slopes with greater uncertainty. Conclusion Elevated tau‐PET signal occurs ∼2.5 years after estimated age of Aß onset in Braak stages I‐III and ∼10‐12.5 years in temporal‐meta and Braak stages IV‐V, with uncertainty within these groupings. These estimates for later Braak stages are consistent with results from previous work on the timing of positive tau PET visual reads. These findings may have implications for determining the optimal time to administer Aß immunotherapies in asymptomatic individuals.