Peripheral risk factors and their role in biomarker-based screening for dementia in the community

Abstract

Introduction: Peripheral risk factors (PRFs) may correlate with dementia plasma biomarkers, potentially reflecting peripheral rather than brain health. This study explores the associations between PRFs and plasma biomarkers glial fibrillary acidic protein (GFAP), neurofilament light chain (NfL), and total-tau, and their role in predicting future dementia.

Methods: Data from the Age, Gene/Environment Susceptibility-Reykjavik Study (2002-2015) included 4353 participants mean age of 76.6 years. A subsample of 910 participants tested their association with PRFs and plasma biomarkers' predictive performance. Sociodemographic, clinical, laboratory, sensory, and lifestyle variables (n = 305) were grouped into 34 clusters.

Results: Besides age and estimated glomerular filtration rate (eGFR), significant associations were found between plasma biomarkers and clusters related to hemoglobin, red blood cell distribution, and inflammation. Incorporating these clusters into predictive models enhanced precision and sensitivity, though overall prediction improvement was modest (area under the precision-recall curve: GFAP 0.17 to 0.34, NfL 0.20 to 0.38).

Discussion: PRFs are significantly associated with dementia plasma biomarkers; Considering these factors may enhance the predictive accuracy of dementia biomarkers.

Highlights: Machine learning identifies key peripheral factors influencing neurodegenerative biomarkers. Hemoglobin and red blood cell distribution cluster associates significantly with biomarker levels. Incorporating diverse peripheral factors modestly enhances incident dementia prediction accuracy in community settings.

Keywords: cluster; incident dementia; peripheral health factors; plasma biomarker; predictive modeling.