Brain Amyloid Plaque Levels Affect Clinical Progression in Alzheimer Disease: Assessment of Amyloid PET and Change in CDR-SB Utilizing Semi-Mechanistic Model.

Lecanemab is a humanized IgG1 monoclonal antibody binding with high affinity to protofibrils of amyloid-beta (Aβ) protein. In clinical studies, lecanemab has been shown to reduce amyloid markers in early symptomatic Alzheimer's disease and slow decline on clinical endpoints of cognition and function. Nonlinear mixed-effects modeling assessed the correlation between amyloid PET and change in CDR-SB using data from lecanemab phase 2 study (Study 201) and phase 3 study (Study 301; Clarity AD). Data from placebo-treated subjects were used to establish a disease-progression model; the effect of amyloid reduction on disease progression was defined using data from lecanemab-treated subjects. CDR-SB scores were used with beta regression to fit a Richard's function parameterized in terms of baseline CDR-SB, intrinsic rate of disease progression, shape, and precision of the beta distribution. Simulations were conducted to evaluate the impact of lecanemab treatment over 4 years. Baseline CDR-SB was predicted by diagnosis and baseline mini-mental state examination (BMMSE) score. Intrinsic rate of disease progression was predicted by amyloid PET and BMMSE. Amyloid PET was a better predictor of drug effect than lecanemab exposure, demonstrating amyloid reduction as a surrogate marker of efficacy. Simulations projected the difference in CDR-SB between lecanemab and placebo treated subjects continued increasing over 4 years. Patients with low baseline amyloid and less severe disease were projected to have slower disease progression and better outcomes with lecanemab treatment.