Blood Biomarkers for the Diagnosis of Neurodegenerative Dementia: A Systematic Review

Abstract

Importance Accurately diagnosing neurodegenerative dementia is often challenging due to overlapping clinical features. Disease specific biomarkers could enhance diagnostic accuracy. However, CSF analysis procedures and advanced imaging modalities are either invasive or high-priced, and routinely unavailable. Easily accessible disease biomarkers would be of utmost value for accurate differential diagnosis of dementia subtypes. Objective To assess the diagnostic accuracy of blood-based biomarkers for the differential diagnosis of AD from Frontotemporal Lobar Degeneration (FTLD), or AD from Dementia with Lewy Bodies (DLB). Methods Systematic review. Three databases (PubMed, Scopus, and Web of Science) were searched. Studies assessing blood-based biomarkers levels in AD versus FTLD, or AD versus DLB, and its diagnostic accuracy, were selected. When the same biomarker was assessed in three or more studies, a meta-analysis was performed. QUADAS-2 criteria were used for quality assessment. Results Twenty studies were included in this analysis. Collectively, 905 AD patients were compared to 1262 FTLD patients, and 209 AD patients were compared to 246 DLB patients. Regarding biomarkers for AD versus FTLD, excellent discriminative accuracy (AUC >0.9) was found for p-tau181, p-tau217, synaptophysin, synaptopodin, GAP43 and calmodulin. Other biomarkers also demonstrated good accuracy (AUC = 0.8-0.9). For AD versus DLB distinction, only miR-21-5p and miR-451a achieved excellent accuracy (AUC >0.9). Conclusion Encouraging results were found for several biomarkers, alone or in combination. Prospective longitudinal designs and consensual protocols, comprising larger cohorts and homogeneous testing modalities across centres, are essential to validate the clinical value of blood biomarkers for the precise etiological diagnosis of dementia.