Researchers at the Keck School of Medicine of USC report that a blood-based screening algorithm substantially reduced the number of unnecessary PET scans required to recruit participants for a large Alzheimer's prevention trial. The findings were published in Alzheimer's & Dementia: The Journal of the Alzheimer's Association.
The trial in question, AHEAD 3-45, is an international phase 3 study asking whether treating people earlier with lecanemab produces better outcomes. Lecanemab clears the sticky amyloid-beta aggregates linked to Alzheimer's progression and can slow clinical decline by roughly 30%, but it is currently prescribed only after cognitive symptoms have appeared. AHEAD 3-45 goes further back in the disease timeline, enrolling people whose brains show amyloid buildup but whose thinking and memory are still intact - a stage that can precede symptoms by decades.
That design creates a recruitment problem. Only about 30% of cognitively healthy adults over 65 carry enough amyloid to qualify, so trial sites had to screen large numbers of people who looked and felt perfectly well. Before plasma screening was added, the process could stretch to three months from a candidate's first visit to enrollment, and more than 70% of people who underwent PET imaging - the expensive gold-standard scan for detecting amyloid - were told afterward that they did not qualify.
Using a blood test to triage candidates before imaging changes that math. Fewer wasted scans means lower cost per enrolled participant, less time and inconvenience for volunteers, and faster progress toward answers about whether intervening before symptoms emerge actually helps. The same logic could eventually apply well beyond research settings, wherever amyloid status needs to be established without sending everyone through a PET scanner.