# Integrative analysis of AT classification, plasma biomarkers, and cognitive trajectories across diverse dementia syndromes

**Authors:** Jihwan Yun, Min Young Chun, Henrik Zetterberg, Kaj Blennow, Fernando Gonzalez-Ortiz, Nicholas J. Ashton, Daeun Shin, Soyeon Yoon, Heejin Yoo, Jun Pyo Kim, Hongki Ham, Yuna Gu, Hee Jin Kim, Seung Hwan Moon, Hanna Cho, Jae Yong Choi, Byung Hyun Byun, Su Yeon Park, Jeong Ho Ha, Duk L. Na, Sang Won Seo, Hyemin Jang

PMC · DOI: 10.1038/s41598-025-30117-y · 2025-12-11

## TL;DR

This study compares how amyloid and tau brain changes, blood markers, and cognitive decline differ across types of dementia.

## Contribution

The study reveals distinct biomarker and cognitive patterns in Alzheimer's, vascular, and frontotemporal dementia across amyloid-tau stages.

## Key findings

- SVCI and FTD show more advanced clinical stages than ADCI at similar amyloid-tau stages.
- SVCI and FTD have higher plasma biomarkers and faster cognitive decline than controls in early amyloid-tau stages.
- ADCI shows faster decline than controls only in later amyloid-tau stages.

## Abstract

This study aimed to examine amyloid-β (Aβ) and tau (AT) biological stages, plasma biomarkers, and cognitive trajectories according to AT stages in Alzheimer’s disease-related cognitive impairment (ADCI), subcortical vascular cognitive impairment (SVCI), and frontotemporal dementia (FTD). A total of 275 participants (42 cognitively unimpaired [CU], 132 ADCI, 73 SVCI, and 28 FTD) underwent Aβ and tau positron emission tomography for assessment of AT stages. Participants with cognitive impairment (ADCI, SVCI, and FTD) were classified according to clinical stages of mild cognitive impairment or dementia. Plasma biomarkers were analysed, and cognitive trajectories were assessed using mixed-effects models over 6.1 years. SVCI and FTD showed more advanced clinical stages than ADCI at equivalent AT stages. SVCI participants had higher plasma glial fibrillary acidic protein (P = 0.012) and neurofilament light chain (P = 0.025) than CU participants in the A − T− stage. In the A − T− stage, SVCI (β = -0.738, P = 0.002) and FTD (β = -4.016, P < 0.001) showed faster cognitive decline than CU, but not ADCI. In the A + T − stage, ADCI (β = -0.634, P = 0.005) and SVCI (β = -0.690, P = 0.006) showed faster decline than CU. In the A + T + stage, only ADCI exhibited significantly faster decline than CU (β = -1.856, P = 0.008). Distinct plasma biomarker profiles and cognitive trajectories characterise ADCI, SVCI, and FTD across AT classification, highlighting the heterogeneity in pathophysiological mechanisms across dementia types.

The online version contains supplementary material available at 10.1038/s41598-025-30117-y.

## Linked entities

- **Proteins:** MAPT (microtubule associated protein tau)
- **Diseases:** frontotemporal dementia (MONDO:0010857)

## Full-text entities

- **Genes:** GFAP (glial fibrillary acidic protein) [NCBI Gene 2670] {aka ALXDRD}, MAPT (microtubule associated protein tau) [NCBI Gene 4137] {aka DDPAC, FTD1, FTDP-17, MAPTL, MSTD, MTBT1}, APP (amyloid beta precursor protein) [NCBI Gene 351] {aka AAA, ABETA, ABPP, AD1, APPI, CTFgamma}
- **Diseases:** FTD (MESH:D057180), dementia (MESH:D003704), ADCI (MESH:D003072)

## Figures

6 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12775461/full.md

---
Source: https://tomesphere.com/paper/PMC12775461