To elucidate and quantify the mechanisms underlying the canonical sequence from beta-amyloid (A) to tau (T) to neurodegeneration (N) to clinical outcomes in Alzheimer’s disease, and to simulate the potential downstream effects of anti-amyloid therapy.

Published data and biologically plausible mechanisms were used to construct, calibrate, and validate the Q-ATN. The Q-ATN has five modules: 1) anti-amyloid effect, 2) amyloid PET, 3) tau PET, 4) cortical thickness (CT), and 5) cognitive impairment (CDR-SB) and four linkages that quantify the relationships between sequential modules. Treatment with a hypothetical anti-amyloid antibody (vs. placebo) was simulated with the Q-ATN model over 5 years.

Linkage 1 quantifies amyloid dynamics with a concentration-dependent first-order rate constant (kDE) for antibody-mediated plaque removal and a parabolic growth function. Linkage 2 quantifies the effect of amyloid on tau PET based on data from the HABS study (Johnson 2020) and hypothesized mechanisms of tau production and clearance. Linkage 3 quantifies the effect of tau on the rate of change of CT (dCT/dt) using a linear function based on data from La Joie 2020. Linkage 4 quantifies the relationship between medial temporal CT and CDR-SB using a non-linear function based on data from Dickerson 2009.

During placebo treatment, the 5-year simulation shows steady accumulation of amyloid PET and tau PET, a progressive decrease in CT, and a progressive increase in CDR-SB.  In contrast, antibody-mediated amyloid removal leads to a reduction in tau PET, slows cortical thinning, and decreases the rate of cognitive decline.

The Q-ATN model provides a novel quantitative framework that connects anti-amyloid therapy to the amelioration of cognitive decline.