Study Reveals Structural Brain Subgroups in Aging, Key for Early Disease Detection

A recent cohort study has revealed significant insights into the structural brain changes associated with aging in individuals without diagnosed cognitive impairment (WODCI). The study, published in JAMA Psychiatry, utilized a deep learning semi-supervised clustering method leveraging generative adversarial networks to identify distinct subgroups among participants aged 45 to 85 years.

“The 3 subgroups showed distinct associations with [cardiovascular risk factors (CVRFs)], genetics, and subsequent cognitive decline,” researchers wrote in the study. “These subgroups likely reflect multiple underlying neuropathologic processes and affect susceptibility to Alzheimer disease.”

Data was acquired from 1999 to 2020, analyzed from July 2021 to February 2023, and drawn from the Imaging-Based Coordinate System for Aging and Neurodegenerative Diseases (iSTAGING) international consortium. Researchers examined associations with genetics, cardiovascular risk factors (CVRFs), amyloid β (Aβ), and future cognitive decline.

Related: Hearing Loss Prevention Could Delay, Prevent Dementia Onset

The study encompassed 27,402 participants (mean [SD] age, 63.0 [8.3] years; 15,146 female [55%]). Three subgroups were identified: a typical aging subgroup (A1) characterized by modest atrophy and white matter hyperintensity (WMH) load, and 2 accelerated aging subgroups (A2 and A3) exhibiting more pronounced characteristics, particularly beyond age 65.

Subgroup A2 displayed associations with hypertension, white matter hyperintensity (WMH), and genetic variants related to vascular diseases, while subgroup A3 exhibited severe, widespread atrophy and greater cognitive decline. Genetic analyses revealed distinct patterns of association with Alzheimer disease and white matter integrity measures across the subgroups.

These findings, researchers say, suggest that structural brain subgroups in aging individuals without cognitive impairment are linked to cardiovascular risk factors, genetics, and subsequent cognitive decline trajectories. Results could help to “pav[e] pathways toward patient stratification at early asymptomatic stages and promot[e] precision medicine in clinical trials and health care,” researchers noted. 

Reference

Skampardoni I, Nasrallah IM, Abdulkadir A, Wen J, et al. Genetic and Clinical Correlates of AI-Based Brain Aging Patterns in Cognitively Unimpaired Individuals. JAMA Psychiatry. 2024 Feb 14:e235599. doi: 10.1001/jamapsychiatry.2023.5599. Epub ahead of print. PMID: 38353984; PMCID: PMC10867779.