Blood Biomarkers in Frontotemporal Dementia: Review and Meta-Analysis

Abstract

Biomarkers in cerebrospinal fluid (CSF) are useful in the differential diagnosis between frontotemporal dementia (FTD) and Alzheimer's dementia (AD), but require lumbar puncture, which is a moderately invasive procedure that can cause anxiety to patients. Gradually, the measurement of blood biomarkers has been attracting great interest. Testing blood instead of CSF, in order to measure biomarkers, offers numerous advantages because it negates the need for lumbar puncture, it is widely available, and can be repeated, allowing the prediction of disease course. In this study, a systematic review of the existing literature was conducted, as well as meta-analysis with greater emphasis on the most studied biomarkers, p-tau and progranulin. The goal was to give prominence to evidence regarding the use of plasma biomarkers in clinical practice.

Keywords: blood biomarkers; frontotemporal dementia; progranulin.

Figure 1

Flow diagram of collecting and evaluating research used in this review.

Figure 2

Meta-analysis (using the random effects model with DerSimonian–Laird method) of studies measuring plasma progranulin levels in carriers of GRN mutations vs. FTD with no GRN mutation or controls, with subgroup analysis (a). In the forest plot, the left side favors lower levels in the presence of GRN mutations, while the right side favors lower levels in the absence of GRN mutations or in controls. Effect sizes are represented as green squares with 95% confidence intervals. In the summary rows, the weighted average effect (or “combined” effect size) is represented as a diamond (b). A funnel plot of all 11 studies is included [30,37,38,39,41,42,43,44,45,46,47].

Figure 3

Meta-analysis (using the random effects model with DerSimonian–Laird method) of studies measuring p-Tau for all studies included (a) and after removal of the fourth study (triad cohort [50]) (b). In the forest plot, the right side favors higher levels of p-Tau in AD, while the left side favors higher values in FTD. Effect sizes are represented as green squares with 95% confidence intervals. In the summary row, the weighted average effect (or “combined” effect size) is represented as a diamond. (c) Funnel plot of all four studies (all studies are located within the “funnel”) [49,50].