PET-measured amyloid beta accumulates at an accelerated rate in Down syndrome compared to neurotypical populations

Abstract

Introduction: Individuals with Down syndrome (DS) have a high prevalence of Alzheimer's disease (AD) and reveal an earlier age of amyloid beta (Aβ) onset compared to sporadic AD. Differences in amyloid accumulation rates between DS and sporadic AD populations have not been established.

Methods: Participants with ≥ 3 [C-11]PiB scans (spanning > 6 years) and transitioning to Aβ+ were included, resulting in 20 DS and 23 neurotypical (NT) participants. Amyloid accumulation was compared using global standardized uptake value ratio (SUVR) for Aβ deposition, with individual growth rates (r) estimated using the logistic growth model ( $SUVR\left(t\right)=SUV{R}_{BL}+\frac{K}{1+e^{-r\left(t-t_{50}\right)}}$ ).

Results: The average growth rate in the DS cohort was 0.28 (0.08)/year versus 0.20 (0.08)/year for NT ( $p=.002$ ), an increase of 40%.

Discussion: Using individual longitudinal analyses, accelerated amyloid accumulation in DS is observed, This has important considerations for informing treatment trial design and monitoring beta-amyloid changes in future AD studies involving individuals with DS.

Highlights: Aβ accumulation rate was estimated using a logistic growth model. There was no overlap in the age of amyloid positivity between DS and NT cohorts. Participants with DS accumulate amyloid 40% faster than those with sporadic AD.

Keywords: Alzheimer's disease; PET imaging; Trisomy 21; [C‐11]PiB; amyloid beta (Aβ) plaques; longitudinal.

FIGURE 1

Expected trajectory of amyloid accumulation in brain over time as measured by positron emission tomography imaging.

FIGURE 2

Amyloid trajectory plots for Down syndrome (warm colors) and neurotypical (cool colors) participants included in this analysis. Note that individual trajectories have different shades of warm or cool colors, so they are more easily distinguished from those of other participants. Individual participants are plotted piecewise linear (left) and using the logistic growth model (right) with the estimated r.

FIGURE 3

Box plots showing distribution of estimated growth rates represented by r in logistic growth model.

FIGURE 4

Average amyloid trajectories of Down syndrome and neurotypical populations measured using carbon‐11 Pittsburgh compound B normalized to t ₅₀ = 0.

FIGURE 5

Trajectory plots for neurotypical (left) and Down syndrome (right) populations before (top) and after time normalization (bottom) using the Sampled Iterative Local Approximation (SILA) model. The SILA normalized data were then fit with the logistic growth model shown in this figure with the black line.

FIGURE 6

The growth rate of the Down syndrome and neurotypical cohorts with a higher carrying capacity used in the logistic growth model, with 95% confidence intervals represented by the dashed lines and with an increasing upper limit on the model the growth rate decreased for both populations, but at a similar rate, so there was not an overlap between confidence intervals for any of the carrying capacity values observed.