Plasma Proteomic Profiles Predict Future Dementia in Healthy Adults

Medical Laboratory Science Neurology Biochemistry Life Sciences Immunology Medicine
plasma proteome dementia prediction GFAP neurofilament light chain GDF15 vascular dementia

Differential Proteins in Plasma

Plasma Proteomics Predicts the Future Risk of Dementia in Healthy Adults

Background and Significance

Predicting dementia has always been a significant challenge in the medical field. With the development of proteomics, biomarkers in the blood offer new opportunities for predicting the onset of dementia. This study focuses on data from the UK Biobank, involving 52,645 adults without dementia and a follow-up period of 14.1 years, examining the relationship between 1,463 plasma proteins and various types of dementia. The study found that GFAP, NEFL, GDF15, and LTBP2 are highly correlated biomarkers with the occurrence of events. This research provides important guidance for future population dementia risk screening and early intervention.

Study Source

This study was jointly completed by Yu Guo, Jia You, Yi Zhang, Wei-Shi Liu, Yu-Yuan Huang, Ya-Ru Zhang, Wei Zhang, Qiang Dong, Jian-Feng Feng, Wei Cheng, and Jin-Tai Yu. The researchers are all from the Department of Neurology, Huashan Hospital, Fudan University, as well as the National Center for Disease Control and Prevention, National Key Laboratory of Medical Neurobiology, and the Advanced Research Institute for Brain Science. The research results were published in Nature Aging, Volume 4, pages 247-260, February 2024.

Study Process and Details

The study process involved multiple steps of screening and analysis:

  1. Identification of Relevant Proteins: Correlation investigation of 1,463 plasma proteins with dementia.
  2. Importance Ranking: Ranking dementia-related proteins based on their predictive contribution under Model 1 and Model 2.
  3. Evaluation of Prediction Accuracy: Assessing the prediction accuracy for future dementia of the selected proteins using ten-fold cross-validation.
  4. Analysis of Clinical Symptom Risk: Exploring the relationship between high and low baseline levels of top-ranked proteins and the progression of clinical symptoms.
  5. Verification of Result Robustness: Randomly dividing the study population into two subsets for repeated analyses to confirm the stability of results.

Research Findings

The changes in GFAP, NEFL, and GDF15 started at least 10 years before dementia diagnosis. Patients with higher GFAP levels were 2.32 times more likely to develop dementia in the future. Combining GFAP or GDF15 with demographic characteristics provided good prediction for all-cause dementia (ACD), Alzheimer’s disease (AD), and vascular dementia (VAD).

Conclusion and Research Value

The study identifies GFAP as an optimal biomarker for predicting dementia, demonstrating great scientific and practical value. It offers new perspectives for screening high-risk dementia populations and early intervention. Moreover, the innovative research methods and procedures provide important references for similar future studies.

Research Highlights

  • The results from internal leave-one-region-out cross-validation and random splitting re-validation demonstrate high robustness.
  • The simplicity and accuracy of the predictive model offer significant advantages for clinical applications.
  • The study reveals that GFAP is not only specific in dementia prediction but could also serve as an important indicator for monitoring therapeutic efficacy.