AI and New Drug Targets Advance Alzheimer's Research Pipeline

Artificial intelligence is increasingly used to analyze large, multimodal Alzheimer's datasets and inform target discovery and trial design, as advances in disease biology, large-scale data sharing and AI begin to align. A special issue of the Journal of Prevention of Alzheimer's Disease examines how AI is now being used to support earlier diagnosis of Alzheimer's disease, identify new drug targets and redesign clinical trials.

Commissioned by Gates Ventures and the Alzheimer's Disease Data Initiative, the issue brings together researchers from eight countries and reflects the growing integration of AI methods into Alzheimer's research. According to the Interim Executive Director of the Alzheimer's Disease Data Initiative, the issue reflects the organization's focus on large-scale data sharing and collaboration, bringing together a global coalition of philanthropic, industry, government and nonprofit partners to fundamentally transform Alzheimer's research.

Two FDA-approved disease-modifying treatments are now on the market, and for the first time, simple blood-based diagnostic tests make broad screening a real possibility. The drugs, lecanemab and donanemab, remove the buildup of amyloid plaques in the brain and can "freeze" a person in their current functional state. In 2025, the U.S. Food and Drug Administration accepted the supplemental biologics licensure application for Leqembi Iqlik (lecanemab-irmb), a subcutaneous autoinjector formulation for once-weekly maintenance dosing in the United States.

Indiana University School of Medicine scientists have identified a promising drug target for Alzheimer's disease. The team found that removing an enzyme called IDOL from neurons in the brain substantially reduces amyloid plaques—a hallmark characteristic of the disease—and may provide further resilience against disease progression. The researchers say targeting this enzyme in neurons can be a new way to remove amyloid plaques and improve communication between neurons and lipid metabolism in the brain.

Targeting enzymes in drug development offers key advantages due to their well-defined active sites or 'pockets' where drugs can attach and block their activity. This precision means molecules can be designed that hit the right target with minimal side effects. In the study, published in Alzheimer's & Dementia: The Journal of the Alzheimer's Association, the researchers generated two different animal models of Alzheimer's disease by deleting the IDOL gene in the brain from either within neurons or microglia, the brain's immune cells.

Deletion of IDOL from neurons not only reduced plaques but also reduced levels of apolipoprotein E (APOE), a protein associated with Alzheimer's disease. One of the protein's variants, APOE4, is the strongest risk factor for late-onset Alzheimer's disease. APOE also plays a critical role in lipid metabolism. The team also discovered that levels of receptors that can regulate APOE and amyloid plaques in the brain increased when the enzyme was removed from neurons. These receptors have a critical role in lipid metabolism and healthy neuronal communication. A recent study shows that activating a pathway, which is also regulated by these receptors, provides resilience to cognitive decline in Alzheimer's patients who have high amounts of plaques.

Scientists at Karolinska Institutet in Sweden and the RIKEN Center for Brain Science in Japan have identified two brain receptors that help regulate the breakdown of amyloid beta, the protein that builds up in Alzheimer's disease. Their findings, published in the Journal of Alzheimer's Disease, suggest it may be possible to develop future medications that are both safer and more affordable than today's antibody based treatments.

The research team discovered that two somatostatin receptors, SST1 and SST4, work together to control neprilysin levels in the hippocampus, a region essential for memory. Normally, an enzyme called neprilysin helps clear away amyloid beta. However, neprilysin activity declines with aging and during the progression of the disease. The researchers conducted experiments using genetically modified mice and laboratory grown cells. When both SST1 and SST4 receptors were missing, neprilysin levels dropped. As a result, amyloid beta accumulated and the mice showed memory problems.

The team also tested a compound designed to activate these two receptors. In mice with Alzheimer's-like brain changes, stimulating SST1 and SST4 increased neprilysin levels, reduced amyloid beta buildup, and improved behavior. Importantly, the treatment did not cause serious side effects. SST1 and SST4 belong to a large family of proteins known as G protein-coupled receptors. These receptors are common drug targets because they are well understood and often respond to medications that can be produced at lower cost and taken orally.

Researchers led by the University of California, Irvine's Joe C. Wen School of Population & Public Health have created detailed maps showing how genes causally regulate one another across different types of brain cells affected by Alzheimer's disease. Using their newly developed machine learning framework, SIGNET, which reveals cause-and-effect relationships rather than simple genetic correlations, they uncovered key biological pathways that may drive memory loss and brain degeneration. The study, published in Alzheimer's & Dementia: The Journal of the Alzheimer's Association, also identifies new genes that could serve as targets for future treatments.

To create these maps, team members analyzed single-cell molecular data from brain samples of 272 participants in long-term memory and aging studies in the Religious Orders Study and the Rush Memory and Aging Project. They developed SIGNET as a scalable, high-performance computing method that integrates single-cell RNA sequencing and whole-genome sequencing data and reveals cause-and-effect relationships among all genes. The researchers identified causal gene regulatory networks for six major types of brain cells. The scientists found that the most dramatic gene disruptions in Alzheimer's disease occur in excitatory neurons—the nerve cells that send activating signals—with analyses of nearly 6,000 cause-and-effect interactions indicating that these cells undergo extensive rewiring as the disease progresses.

Clinical development remains highly active, with 138 agents in 182 trials in 2025 spanning non-amyloid biology, including bioenergetics, endocrine pathways, oxidative stress, synaptic plasticity, and vascular mechanisms. Dementia affects more than 55 million people worldwide, with Alzheimer's disease accounting for approximately 70% of cases. By 2050, it is estimated that around 130 million people could suffer from Alzheimer's disease, driven by an ageing global population.

In Korea, the government finalized the fifth Comprehensive Plan for Dementia Management, which covers the period from 2026 to 2030. The plan shifts the focus from expanding physical infrastructure to prioritizing patient rights and highly customized services. Support for AI research, including multimodal foundation models for brain cognitive function analysis, will be increased to develop early diagnosis and personalized treatments. Following the commercial launch of Leqembi (lecanemab) in late 2024, industry sources confirm that Eli Lilly Korea submitted a product approval application to the Ministry of Food and Drug Safety for its Alzheimer's treatment, Kisunla (donanemab), in late 2025.