Children of parents with dementia showed an older-looking immune system, but not faster aging overall

Having a parent with dementia raises a person's own risk of Alzheimer's disease. Part of that is shared genes and part is shared surroundings and habits, but how the risk shows up in the body years before any memory problems is not well understood. A team at Columbia University asked a simple question: do the middle-aged children of people with dementia already look biologically older?

Mostly, they did not. Their epigenetic clocks and brain scans looked like everyone else's. What differed was their immune system: they had more of a type of white blood cell that builds up as the immune system ages. The study is an observational study and a preprint, not yet peer-reviewed.

Some background. Epigenetic clocks read chemical tags on DNA (DNA methylation) in a blood sample and turn them into an estimate of biological age, or of how fast someone is aging. The same tags can also be used to estimate which kinds of white blood cells are in the sample. Among these are T cells, the immune cells that recognise and kill infected or abnormal cells. "Naive" T cells have not yet met their target; "memory" T cells have. As people age, the balance shifts from naive towards memory cells, so a memory-heavy profile is one sign of an aging immune system.

What they did

• Who: 1,204 adults aged 28 to 89 (average about 56) in the Offspring Study in New York City. Most were Hispanic (72%) or Black (21%), and two thirds were women. Their parents had been followed in a long-running aging study at Columbia, where doctors diagnosed dementia using tests and exams, so the parents' status was not just reported by the children.
• Groups: 427 had a parent who developed dementia during that follow-up; 776 did not. Those with an affected parent were older on average (about 60 versus 54) and more often had diabetes, so the analyses took age and sex into account.
• Measurements: one blood sample from each person, used to calculate several epigenetic clocks, an estimate of telomere length (telomeres are the protective caps at the ends of chromosomes) and the share of different immune cells. In 680 people, an MRI scan was used to estimate "brain age".
• Because they tested many measures at once, the team applied a statistical correction so that chance findings would not pass as real ones.

What they found

• No faster aging by the clocks. After the correction, people with a parent with dementia did not differ on any epigenetic clock, nor on the two MRI-based measures of brain aging.
• More memory killer T cells. They had a higher estimated share of memory CD8 T cells, the "killer" type. This was the only difference that survived the correction. Smaller shifts in the same direction (more memory CD4 T cells and memory B cells, fewer naive CD4 T cells) did not.
• Shorter estimated telomeres, explained by the T cells. Telomeres looked slightly shorter in this group, but once the share of memory CD8 T cells was taken into account, that difference disappeared.
• When the parent got sick, and how sharp the parent was. Among the 331 people whose parent's age at diagnosis was known, an earlier diagnosis went with a slightly faster pace of aging on one clock (DunedinPACE). Across everyone, children of parents who scored better on memory and thinking tests had a slower pace of aging, longer estimated telomeres, a younger-looking brain on MRI and a more naive-leaning immune profile.

What to keep in mind

• Association, not cause. The authors note that the study can't tell how risk passes from parent to child: shared genes, environment and lifestyle may all play a part.
• Estimated, not counted. The immune cell shares were calculated from DNA tags, not counted directly in the blood. The authors call these results exploratory and say direct counts would be needed to confirm them. The DNA-based telomere estimate also agrees only weakly with the standard lab test.
• Small effects. The differences were modest, and the share of the telomere link "explained" by T cells has a very wide margin of error, which the authors say should be read with caution.
• One city, one moment. All participants lived in New York City, which may limit how far the results apply elsewhere, and each person gave one blood sample, so the study shows a snapshot, not change over time.
• Not yet peer-reviewed.

Why it matters

Children of people with dementia are a group that researchers would like to follow early, long before any symptoms. This study suggests that, in midlife, their family risk does not show up as a body that is aging faster across the board. Instead it may show up in one place: an immune system tilted towards memory killer T cells, a pattern that other studies have also seen in people with early memory problems and Alzheimer's disease. Whether that tilt is part of how dementia develops, or simply travels with it, is a question this kind of study can raise but not answer.