# Neuropathologic Profiles and Associated Cognitive Trajectories in Community-Living Older Adults

## TL;DR

This study identifies five distinct brain disease profiles in older adults and shows how each profile is linked to different patterns of cognitive decline over time.

## Contribution

The study introduces a new method to classify complex brain pathologies and links them to specific cognitive aging trajectories.

## Key findings

- Five distinct neuropathologic profiles were identified in older adults with mixed brain pathologies.
- Profiles with high ADNC or LATE-NC and hippocampal sclerosis showed the strongest associations with cognitive decline.
- Cognitive decline varied in timing and rate across the five neuropathologic profiles.

## Abstract

This cohort study describes distinct neuropathologic profiles in the donated brains of older adults and examines associated cognitive trajectories over time..

What are the latent neuropathologic profiles and associated cognitive trajectories in community-living older adults?

In this cohort study with 1633 older adults, more than 80% of the individuals had mixed pathologies, with 280 distinct combinations of copathologies. Hierarchical clustering and functional mixed-effects models identified 5 distinct neuropathologic profiles and associated cognitive trajectories.

These findings suggest that mixed neuropathologies are complex, some profiles have stronger associations with cognitive aging.

Mixed neuropathologies are common, and yet the full extent to which these pathologies coexist is not entirely clear.

This study aims to identify distinct neuropathologic profiles in community-dwelling older adults and to examine associated cognitive trajectories over time.

This study included participants in 2 community-based cohort studies of aging and dementia. Participants were older adults without known dementia at enrollment who agreed to annual evaluations and brain donation. The Religious Orders Study started in 1994, and the Rush Memory and Aging Project started in 1997. Both studies are ongoing. Data were analyzed in May 2025.

Annual uniform detailed evaluations for up to 30 years and brain autopsy and neuropathologic evaluation after death.

Neuropathologic evaluations assessed Alzheimer disease neuropathologic change (ADNC), Lewy bodies, limbic-predominant age-related transactive response DNA-binding protein 43 kDa encephalopathy neuropathologic change (LATE-NC), hippocampal sclerosis, infarcts, cerebral amyloid angiopathy, atherosclerosis, and arteriolosclerosis. Annual cognitive scores were derived from a battery of 19 tests that assessed multiple cognitive domains. Latent neuropathologic profiles were identified using hierarchical clustering, and longitudinal cognitive trajectories associated with each profile were estimated using functional mixed-effects models.

A total of 1633 older adults were included in the study. The mean (SD) age at death was 90.4 (6.4) years, 1156 (70.8%) were female, and the mean (SD) years of education was 16.2 (3.6) years. A total of 46 participants (2.8%) self-reported as Black, 42 (2.6%) as Hispanic, and 1579 (96.6%) as White. More than 80% of individuals had mixed neuropathologies at autopsy, and 280 unique combinations of copathologies were identified. Hierarchical clustering revealed 5 distinct neuropathologic profiles: profile 1 (259 participants [15.9%]) was characterized by a high burden of infarcts and vessel diseases, profile 2 (201 participants [12.3%]) by high LATE-NC and hippocampal sclerosis, profile 3 (355 participants [21.7%]) by high Lewy bodies, profile 4 (159 participants [9.7%]) by high ADNC and amyloid angiopathy, and profile 5 (659 participants [40.4%]) by low pathology overall. Cognitive trajectories differed across profiles in terms of both the rate of decline and the timing of the onset of decline with fastest decline observed for profile 2 and 4.

In this cohort study, mixed neuropathologies were extremely common and complex. Compared with vascular conditions, degenerative pathologies clustered into distinct profiles. Profiles characterized by high burdens of ADNC and separately, LATE-NC and hippocampal sclerosis, had the most potent associations with cognitive decline.