Scenarios for the long-term efficacy of amyloid-targeting therapies in the context of the natural history of Alzheimer’s disease

Importance Recent clinical trials of Aβ-targeting therapies in Alzheimer’s disease (AD) have demonstrated clinical benefit over 18-months, but their long-term impact on disease trajectory is not yet understood. We propose a framework for evaluating realistic long-term scenarios. Objective To integrate data from recent phase 3 trials of the high-clearance Aβ-targeting antibodies with an estimate of the long-term patient-level natural history trajectory of the Clinical Dementia Rating-Sum of Boxes (CDR-SB) score, to explore realistic long-term efficacy scenarios. Data Sources Published results from recent phase 3 randomized clinical trials of high-clearance Aβ-targeting antibodies with completion between 2019 and 2023. Study Selection All phase 3 studies of aducanumab, lecanemab, gantenerumab, and donanemab were included. Main Outcomes and Measures Reported study results on CDR-SB estimated using categorical-time models. Results Three distinct long-term efficacy scenarios were examined, ranging from conservative (enduring short-term delay), over intermediate (fading stage-dependent slowing), to optimistic (continued stage-independent slowing). In a hypothetical framework, we found that even modestly effective therapies with fading stage-dependent slowing could result in major delays of disease progression if initiated in the earliest stages of AD. In long-term scenarios for the treatment effects observed in recent positive phase 3 trials, we found that initiating the treatments in the early symptomatic stages, defined by the respective trial inclusion criteria, could delay the onset of severe dementia by 0.3-0.6 years (conservative), 1.1-1.9 years (intermediate), and 2.0-4.2 years (optimistic). Conclusion and Relevance The findings provide insights into hypothetical long-term impact of Aβ-targeting treatments, highlighting the potential of maximizing clinical benefit with earlier intervention. This study underscores some of the complexities of evaluating and comparing Aβ-targeting therapies in the context of AD’s nonlinear disease trajectory and the need for considering both differences in trial populations and duration. Our work calls for studies with longer follow-up and results from early intervention trials to provide a comprehensive assessment of these therapies’ true long-term impact.