Targeted blood proteome profiling using NULISAseq identifies

Authors

Wenyue Zheng, Yuanbing Jiang, Hiu Yi Wong, Wan Wa Wong, Lily K W Cheng, Elaine Y L Cheng, Bonnie W Y Wong, Ronnie M N Lo, Siu Ki Leung, Fanny C Ip, Jacqueline K Y Yuen, Yat Fung Shea, Wai Ming Wong, Chun Keung Shum, Hok Man Wai, Vincent C T Mok, Timothy C Y Kwok, Kin Y Mok, Amanda Heslegrave, John Hardy, Henrik Zetterberg, Amy K Y Fu, Nancy Y Ip

Abstract

Alzheimers Dement. 2026 Feb;22(2):e71179. doi: 10.1002/alz.71179.

ABSTRACT

INTRODUCTION: While current blood-based biomarkers for Alzheimer's disease (AD) are effective for determining amyloid beta (Aβ) pathology positivity/negativity, they are insufficient for quantifying Aβ plaque deposition.

METHODS: We profiled 325 plasma proteins in a Hong Kong Chinese cohort using the Nucleic Acid Linked Immuno‑Sandwich Assay (NULISAseq) platform. We analyzed the dysregulation trajectories of the blood proteome along Aβ pathology progression and used machine learning to develop a biomarker panel to quantify Aβ pathology.

RESULTS: We identified 43 blood proteins correlated with Aβ plaque accumulation and selected 8 proteins to construct a model. This model was strongly correlated with amyloid positron emission tomography Centiloid values (r = 0.89), enabling quantification of Aβ deposition and classification of early-stage pathology (area under the curve = 0.93).

DISCUSSION: This study provides a systematic profile of dynamic protein alterations during Aβ pathology progression. Moreover, we developed a biomarker assay that accurately quantifies Aβ pathology, offering a potential tool to facilitate early screening and monitoring of amyloid pathology.

PMID:41612914 | DOI:10.1002/alz.71179