Plasma amyloid-β42/40 and apolipoprotein E for amyloid PET pre-screening in secondary prevention trials of Alzheimer's disease

The extent to which newly developed blood-based biomarkers could reduce screening costs in secondary prevention trials of Alzheimer's disease is mostly unexplored. We collected plasma amyloid-beta 42/40, apolipoprotein E epsilon 4 status and amyloid PET at baseline in 181 cognitively unimpaired participants [the age of 72.9 (5.3) years; 61.9% female; education of 11.9 (3.4) years] from the Swedish BioFINDER-1 study. We tested whether a model predicting amyloid PET status from plasma amyloid-beta 42/40, apolipoprotein E status and age (combined) reduced cost of recruiting amyloid PET + cognitively unimpaired participants into a theoretical trial. We found that the percentage of cognitively unimpaired participants with an amyloid PET + scan rose from 29% in an unscreened population to 64% [(49, 79); P < 0.0001] when using the biomarker model to screen for high risk for amyloid PET + status. In simulations, plasma screening also resulted in a 54% reduction of the total number of amyloid PET scans required and reduced total recruitment costs by 43% [(31, 56), P < 0.001] compared to no pre-screening when assuming a 16x PET-to-plasma cost ratio. Total savings remained significant when the PET-to-plasma cost ratio was assumed to be 8x or 4x. This suggests that a simple plasma biomarker model could lower recruitment costs in Alzheimer's trials requiring amyloid PET positivity for inclusion. Cullen et al. demonstrated that combining plasma amyloid-beta 42/40 with APOE status and age could be used to effectively screen cognitively unimpaired individuals in the Swedish BioFINDER study for amyloid PET abnormality. These results add to the evidence that this panel could greatly improve recruitment for clinical trials.