Cerebrospinal fluid proteomics identification of biomarkers for amyloid and tau PET stages
Zhibo Wang, Yuhan Chen, Katherine Gong, Bote Zhao, Yuye Ning, Meilin Chen, Yan Li, Muhammad Ali, Jigyasha Timsina, Menghan Liu, Carlos Cruchaga, Jianping Jia
Abstract
Accurate staging of Alzheimer’s disease (AD) pathology is crucial for therapeutic trials and prognosis, but existing fluid biomarkers lack specificity, especially for assessing tau deposition severity, in amyloid-beta (Aβ)-positive patients. We analyze cerebrospinal fluid (CSF) samples from 136 participants in the Alzheimer’s Disease Neuroimaging Initiative using more than 6,000 proteins. We apply machine learning to predict AD pathological stages defined by amyloid and tau positron emission tomography (PET). We identify two distinct protein panels: 16 proteins, including neurofilament heavy chain (NEFH) and SPARC-related modular calcium-binding protein 1 (SMOC1), that distinguished Aβ-negative/tau-negative (A−T−) from A+ individuals and nine proteins, such as HCLS1-associated protein X-1 (HAX1) and glucose-6-phosphate isomerase (GPI), that differentiated A+T+ from A+T− stages. These signatures outperform the established CSF biomarkers (area under the curve [AUC]: 0.92 versus 0.67–0.70) and accurately predicted disease progression over a decade. The findings are validated in both internal and external cohorts. These results underscore the potential of proteomic-based signatures to refine AD diagnostic criteria and improve patient stratification in clinical trials. • The early-stage protein panel distinguishes amyloid PET-positive and negative cases • The late-stage protein panel identifies tau PET status in amyloid-positive individuals • The protein panel predicts dementia progression and cognitive decline over 10 years • Identified proteins are implicated in synaptic plasticity and metabolic dysfunction Wang et al. apply machine learning to proteomic profiles to identify proteins predictive of Alzheimer’s disease pathology stages. They develop two models: one for early-stage amyloid positivity and another for assessing tau pathology severity in amyloid-positive patients. Both models can predict dementia progression and cognitive decline over 10 years.