Repeating DNA Stretch in FAM193B Gene Linked to Higher Alzheimer’s Risk

A short, repeating stretch of DNA in the control region of a little-studied gene called FAM193B is linked to higher odds of Alzheimer’s disease, according to a study led by researchers at the Hong Kong University of Science and Technology and published in Nature Neuroscience on Oct. 12, 2026. The work adds a new piece to the puzzle of Alzheimer’s genetic risk, though the authors and outside reporting stress that it shows an association across groups, not a diagnosis for any one person.

Image: Stained cortical neurons under a microscope, used for illustration. Photo: Ns takes photos, via Wikimedia Commons (CC BY 4.0).

What the researchers found

The repeating sequence is three DNA letters, CCG, stacked one after another. Its length varies widely between people, from four copies to more than 30. Because each person carries two copies of the region, the team compared people based on their longer copy.

Key results from the paper and a Medical Xpress summary published Oct. 19:

• The main analysis covered 1,516 people with Alzheimer’s and 1,674 cognitively normal people from a Hong Kong cohort and an independent mainland Chinese cohort.
• People whose longer copy had at least 18 CCG repeats had about twice the odds of Alzheimer’s (odds ratio 2.11) compared with people whose longer copy had 11 or fewer.
• That 18-plus version was common: it appeared in about 19.9% of cognitively normal participants and 30.9% of people with Alzheimer’s in the Chinese cohorts.
• About 5.9% of Chinese individuals carried at least 21 repeats, which was tied to higher odds still (odds ratio 2.89).
• In the UK Biobank, among people of European ancestry (2,674 clinically diagnosed cases and 22,996 controls), the link was smaller, with an odds ratio of about 1.21 for alleles with 19 or more repeats.

For context, the paper notes that the best-known Alzheimer’s risk gene variant, APOE e4, carries an odds ratio of about 3.2 for people with one copy, and that earlier genome-wide studies have found more than 70 risk locations in the genome.

How longer repeats may affect brain cells

Because the repeat sits in the gene’s promoter, the stretch of DNA that acts like a dimmer switch, the team tested whether longer repeats turn the gene up. Blood samples and donated brain tissue showed that longer repeats went along with higher FAM193B activity. In lab experiments, a version with 25 repeats drove more gene activity than one with 11 and bound more strongly to MAZ, a protein that helps regulate genes.

When the researchers boosted FAM193B in neuroblastoma-derived cells, a common lab model for nerve cells, the cells showed more markers of DNA damage and changes in genes tied to stress responses and cell death. In brain scans from the Hong Kong cohort, longer repeats were associated with lower gray matter volume.

The authors acknowledge that the gene’s normal biological function “remains unclear.” Medical Xpress noted that the cell results hint at a connection to cellular harm but do not show that longer repeats damage neurons in people.

Why structural variants matter for Alzheimer’s genetic risk

Most genetic studies of Alzheimer’s look at single-letter differences in DNA. Larger changes, such as insertions, deletions and repeats longer than 50 letters, are harder to read with standard short-read sequencing. The Hong Kong team combined long-read sequencing from 11 Chinese genomes with short-read data from more than 3,000 people to catalog these larger variants. The paper reports that adding them to genetic risk scores explained more Alzheimer’s risk than single-letter variants alone.

The study’s corresponding authors are Amy K. Y. Fu and Nancy Y. Ip of HKUST. Co-authors include John Hardy of University College London’s UK Dementia Research Institute.

What it means for patients and Boston’s research community

The findings do not change testing or treatment today. The researchers say larger studies and more work on what FAM193B does in neurons are needed, and the difference in effect size between Chinese and European participants means the estimates cannot simply be applied across ancestries.

The stakes are large. The Alzheimer’s Association estimates that 7.4 million Americans age 65 and older are living with Alzheimer’s in 2026, about 1 in 9 people in that age group. Greater Boston is a major center for brain-disease research and drug development, a point underscored by Danish brain-health drugmaker Lundbeck’s move of its U.S. headquarters to Greater Boston. Our earlier coverage of a long COVID study on dopamine nerve endings looked at another line of brain research making news this fall.

Sources

• Nature Neuroscience: Genome-wide analysis of common structural variants in Alzheimer’s disease identifies a risk-associated regulatory CCG repeat in FAM193B
• Medical Xpress: Repeating stretch of DNA offers a new clue to Alzheimer’s susceptibility
• Alzheimer’s Association: 2026 Alzheimer’s Disease Facts and Figures
• Wikimedia Commons: Cortical Neurons, stained (CC BY 4.0)