Clinical trajectories are highly variable among patients with Alzheimer’s disease, even when those patients are confirmed to be amyloid-positive. In clinical trials, this dilutes statistical power and biases comparisons of placebo and treatment.
ADMdx has developed imaging classifiers that are predictive of patient clinical trajectory. We have demonstrated these in AD dementia as well as prodromal AD populations, using volumetric MRI, FDG PET, and early frame amyloid imaging.
For example, shown at right is one of our AD Progression Classifiers. The score on the y-axis reflects the degree to which a patient’s scan expresses the pattern of decline shown (in this FDG PET example, blue indicates hypometabolism and red preserved metabolism; we have developed similar classifiers using brain volume, perfusion (early frame amyloid), and tau). Among amyloid-positive late MCI patients from ADNI, those with low baseline classifier scores had stable MMSE scores over the next 36 months, whereas those with higher classifier scores at baseline showed greater subsequent decline in MMSE.