Extended Data Fig. 4: Association between physical activity levels and baseline Aβ burden on longitudinal (a) PACC5 decline and (b) CDR-SOB progression.

Linear mixed effects models revealed interaction between baseline physical activity levels (ordinal) and Aβ burden (continuous) on longitudinal PACC5 decline and CDR-SOB progression. Using the inactive subgroup as reference, all higher levels of physical activity were associated with slower Aβ-related PACC5 decline and CDR-SOB progression, except the slower CDR-SOB progression in the low activity group did not reach statistical significance (Table 3). To visualize the model results, the estimated trajectories based on representative levels of low versus high baseline Aβ burden across physical activity levels are presented, with error bands representing 95% confidence intervals for the estimated trajectories. Low and high Aβ are represented, for illustration purposes, by the mean Aβ burden of Aβ-negative (PiB PVC-DVR 1.17) and Aβ-positive (PiB PVC-DVR 1.85) participants respectively, defined using the conventional Aβ-positivity threshold (PiB PVC-DVR of 1.324). The horizontal dotted line represents thresholds for cognitive impairment (-1.5 PACC5 z-score) and functional decline (1.5 points on CDR-SOB). The vertical dot-dash line represents median duration of cognitive follow-up (9 years). Aβ = β-amyloid; DVR = distribution volume ratio; PiB = Pittsburgh compound-B; PVC = partial volume corrected; SUVR = standardized uptake value ratio.