InterventionsHumans111,680 participantsMendelian randomisation

Genetic data highlight potential drug targets for Alzheimer's

A Mendelian randomisation analysis identified 19 candidate targets in people of European ancestry and seven in people of African ancestry.

Figure 1 from Nature Neuroscience
Open the figure at full size
Figure 1Hou et al.

Nature Neuroscience

In a Mendelian randomisation analysis of genetic data from people with Alzheimer's disease, researchers screened molecular datasets to nominate druggable targets. They identified 19 candidate targets for individuals of European ancestry and seven for individuals of African ancestry. Among these was epoxide hydrolase 2, known as EPHX2. In lab-grown nerve cells and cerebral organoids derived from patients with Alzheimer's disease, a protective genetic variant reduced levels of EPHX2 and phosphorylated tau. In a mouse model of the disease, chemical inhibition of EPHX2 improved cognitive performance. Finally, in an observational analysis of electronic health records from 111,680 patients with mild cognitive impairment, 12 drugs were associated with reduced incidence of Alzheimer's disease, including trazodone and baclofen.

Why it matters

Alzheimer's disease is a major neurodegenerative condition that primarily affects older adults. Pinpointing candidate molecular targets and existing medicines linked to lower incidence bears on efforts to find new ways to treat age-related cognitive decline.

Caveats

The findings on existing medications come from observational health records, which show associations rather than proof of cause and effect. In addition, the experiments in cultured cells and mice cannot ensure that blocking EPHX2 will safely aid cognition in humans.

The paper

Alzheimer's disease target and drug discovery by leveraging multiomics and electronic health data

Yuan Hou, Yichen Li, Pengyue Zhang,
Show 33 more authorsNoah Lorincz-Comi, Dhruv Gohel, Fan Fan, Yunguang Qiu, Jun Yang, Xin Chen, Wenqiang Song, Xiaoyu Yang, Zhibing Tan, Zhigang Liu, Xing Fang, Isabela Rivera Paz, William Martin, Yayan Feng, Yadi Zhou, Jielin Xu, Lijun Dou, Jane Border, Huawei Zhang, Jena' N Mazique, Sung Hee Hwang, Richard J Roman, Tousi Babak, Lynn Bekris, Ehud Karavani, Michael Danziger, Michal Rosen-Zvi, Jonathan L Haines, Haiyuan Yu, Bruce D Hammock, James B Leverenz, Andrew A Pieper, Jeffrey Cummings,
Feixiong Cheng

Cleveland Clinic Genome Center

Nature Neuroscience · 8 Oct 2026

doi.org/10.1038/s41593-026-02472-0PubMed 42850312