Novel fluid biomarkers to differentiate frontotemporal dementia and dementia with Lewy bodies from Alzheimer's disease: A systematic review

Rationale: Frontotemporal dementia (FTD) and dementia with Lewy bodies (DLB) are two common forms of neurodegenerative dementia, subsequent to Alzheimer's disease (AD). AD is the only dementia that includes clinically validated cerebrospinal fluid (CSF) biomarkers in the diagnostic criteria. FTD and DLB often overlap with AD in their clinical and pathological features, making it challenging to differentiate between these conditions. Aim: This systematic review aimed to identify if novel fluid biomarkers are useful in differentiating FTD and DLB from AD. Increasing the certainty of the differentiation between dementia subtypes would be advantageous clinically and in research. Methods: PubMed and Scopus were searched for studies that quantified and assessed diagnostic accuracy of novel fluid biomarkers in clinically diagnosed patients with FTD or DLB, in comparison to patients with AD. Meta-analyses were performed on biomarkers that were quantified in 3 studies or more. Results: The search strategy yielded 614 results, from which, 27 studies were included. When comparing bio-fluid levels in AD and FTD patients, neurofilament light chain (NfL) level was often higher in FTD, whilst brain soluble amyloid precursor protein beta (sAPP beta) was higher in patients with AD. When comparing bio-fluid levels in AD and DLB patients, alpha-synuclein ensued heterogeneous findings, while the noradrenaline metabolite (MHPG) was found to be lower in DLB. Ratios of A beta 42/A beta 38 and A beta 42/A beta 40 were lower in AD than FTD and DLB and offered better diagnostic accuracy than raw amyloid-beta (A beta) concentrations. Conclusions: Several promising novel biomarkers were highlighted in this review. Combinations of fluid biomarkers were more often useful than individual biomarkers in distinguishing subtypes of dementia. Considering the heterogeneity in methods and results between the studies, further validation, ideally with longitudinal prospective designs with large sample sizes and unified protocols, are fundamental before conclusions can be finalised.