Agentic AI Could Reshape Alzheimer's Research, Trials, and Care, Task Force Says
A CTAD task force outlines how agentic AI co-scientists could speed Alzheimer's discovery, make trials more efficient, and personalize care.
Summary
At the 2025 Clinical Trials on Alzheimer's Disease conference, an international task force reviewed how a new class of artificial intelligence, called agentic AI, could transform research and care for Alzheimer's disease and related dementias. These systems can act on their own to combine different types of data, propose and test new ideas, and link biological findings to possible treatments, including repurposing existing drugs. The group also described AI tools that could improve clinical trials by measuring outcomes automatically, picking the participants most likely to respond, and predicting how each person's disease will progress. Beyond research, AI may help with diagnosis, care navigation, and personalized disease management. The task force's priorities included AI-ready data systems, closer teamwork between clinicians, researchers, and computer scientists, AI-optimized trials, and strong ethics and governance rules.
Detailed Summary
Artificial intelligence may have major potential to transform research and care for Alzheimer's disease and related dementias (ADRD), from discovery through clinical trials to everyday care.
This paper is a task force report from the 2025 Clinical Trials on Alzheimer's Disease (CTAD) conference. Experts reviewed recent advances in artificial intelligence, with a focus on agentic AI, meaning systems that can pursue goals autonomously rather than just answer prompts. They also discussed priorities for guiding future AI-driven innovation across the ADRD research and care continuum.
The report describes several emerging applications: - Agentic AI co-scientists can integrate multimodal data, generate and evaluate hypotheses, and connect siloed knowledge. - End-to-end pipelines can link mechanisms to therapeutic design by automating molecular docking and drug repurposing. - Digital wet labs allow in silico experiments. - Agentic tools can harmonize data and infer missing information using digital twins and synthetic data. - In clinical trials, AI may automate outcome assessment, stratify participants precisely, model individual disease trajectories, and predict treatment response. These approaches could improve efficiency through responder enrichment. - Applications are also expanding into diagnosis, care navigation, and personalized disease management.
The task force identified priorities including building AI-ready ecosystems, strengthening collaboration among clinicians, researchers, and computer scientists, using AI to optimize clinical trials, and establishing robust ethical and governance frameworks for responsible adoption.
Several caveats apply: - This is an expert consensus and perspective piece, not a study testing whether these tools improve patient outcomes. - Many applications are described as emerging or potential. - Several authors disclose industry ties. Two authors are co-founders of Vaicro Inc., and one has a pending patent on a multi-agent AI system for neurodegenerative disease diagnosis and discovery. - This summary is based on the abstract and the disclosure statement only.
Key Findings
- Agentic AI co-scientists may autonomously integrate multimodal data, generate hypotheses, and connect siloed Alzheimer's knowledge.
- AI pipelines can automate molecular docking and drug repurposing to link disease mechanisms to candidate therapies.
- Digital twins and synthetic data may fill data gaps and ease pre-discovery bottlenecks like harmonization.
- AI-driven stratification and response prediction could boost trial efficiency through responder enrichment.
- Task force priorities: AI-ready ecosystems, cross-disciplinary collaboration, optimized trials, and ethical governance frameworks.
Methodology
This is a task force report from the 2025 CTAD conference, summarizing expert review of recent AI advances and consensus discussion of priorities. It is not a systematic review or an experimental study.
Study Limitations
The summary is based on the abstract only. The report is expert opinion without presented outcome data showing patient benefit, and many applications are still emerging. Authors disclose industry ties, and some have commercial and patent interests in AI for neurodegeneration.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe:
