A 12-month randomised controlled trial in 81 adults with Down syndrome (mean age 27) has reported that participants offered remotely delivered group exercise three times a week improved on processing speed (estimated marginal mean change 0.17, p = 0.04) and showed a small increase in right hippocampal volume (estimated marginal mean change 0.004, p = 0.04) over the year. Total grey matter volume fell in the arm offered one session a week (estimated marginal mean change -0.82, p = 0.02) and did not change in the three-sessions arm (0.08, p = 0.72), producing a significant group effect across time (0.90, p = 0.04). The paper, by Lauren Ptomey and colleagues at the University of Kansas Medical Center, was published online in Disability and Health Journal on 30 July 2026 and is still listed as in press.
The trial matters because adults with Down syndrome carry the highest genetic risk of Alzheimer's disease: the extra copy of chromosome 21 carries the amyloid precursor protein gene, and the same group's design paper cites a median age of dementia onset of 53.8 years and a cumulative incidence above 90 percent by age 65. Participants were randomised 2:2:1 to three 40-minute remote group sessions a week (n = 34), one a week (n = 32), or a support and education control (n = 15), all with twice-monthly individual support sessions. The trial ran from February 2020 to November 2023, was registered as NCT04048759, and reported 100 percent retention with attendance averaging about 85 percent at exercise sessions. Cognition was measured with the Down syndrome adapted CANTAB battery and brain volumes by MRI at baseline and 12 months.
The counterweights are substantial and the authors report them. There were no significant changes in executive function or episodic memory in any group (all p > 0.05), and no between-group differences in overall cognitive scores; the processing speed and hippocampal results are within-arm changes at p = 0.04 in a small trial with only 15 people in the control arm. The trial's main outcomes paper, published in Alzheimer's and Dementia: Translational Research and Clinical Interventions in June 2025, found no significant difference between arms in change in daily moderate to vigorous physical activity (10 minutes a day in the three-sessions arm, 2 in the once-weekly arm, p = 0.06; 1 in the control arm, p = 0.13), and cardiorespiratory fitness change differed between the two exercise arms but not between the three-sessions arm and the control.
The authors write that the results "suggest that structured exercise may influence cognition and brain health in adults with DS." What the trial establishes firmly is that a year of supervised group exercise can be delivered remotely to this population and kept up: nobody dropped out, and most sessions were attended. Whether that translates into later dementia risk is untested here, and would need a longer trial with clinical or biomarker endpoints.