From proof of principle to therapeutic discipline: strategies to improve the benefit/risk profile of amyloid-targeting treatments for Alzheimer’s disease

The first generation of clinically approved amyloid-targeting treatments has changed the terms of debate in Alzheimer's disease. It is no longer persuasive to argue that amyloid removal is biologically irrelevant or clinically inert in early symptomatic Alzheimer's disease. Lecanemab and donanemab have both demonstrated statistically robust and clinically meaningful, albeit still modest, slowing of disease progression, thereby establishing disease modification as a realistic therapeutic objective rather than a speculative aspiration. At the same time, these agents have exposed the defining limitation of the current therapeutic concept: although clearly beneficial for some patients, the margin for benefit versus risk remains narrow, and its realization in routine care is operationally demanding and heavily dependent on careful patient selection and surveillance rather than on drug effect alone. The central task for the field is therefore no longer to prove that amyloid-targeting treatments can work, but to make them work more effectively, more safely, more targeted, and less burdensome. In current practice, most plausible gains in net clinical value will come from earlier and more precise patient selection, structured risk stratification using the APOE genotype and baseline magnetic resonance imaging status, strict management of vascular risk factors, intensified monitoring during the early hazard period, more personalized treatment exposure, and amyloid antibodies engineered for lower vascular toxicity, for example, using brain shuttle technology. Further, alternative routes of administration have the potential to reduce burden on patients, caregivers, and providers likewise. Real-world registries will be essential to define long-term safety, effectiveness, and treatment patterns outside clinical trial populations. Therefore, future improvement of amyloid-targeting treatments is likely to come from widening the therapeutic margin by drug engineering as well as precision implementation in real-world settings.