AbstractBackgroundFrontotemporal dementia (FTD) is the second most common form of presenile dementia. The differential diagnosis of FTD is still a challenging task due to its symptomatic overlap with other neurological diseases and the lack of diagnostic biomarkers. Here, we aimed to investigate the diagnostic potential of novel biomarkers in cerebrospinal fluid (CSF) and blood.MethodWe included 137 patients from the Centre for Memory Disturbances in Perugia with a diagnosis of FTD (n=37), Alzheimer’s disease (AD, n=47), Parkinson’s Disease (PD, n=14), Dementia with Lewy bodies (DLB, n=8), and cognitively unimpaired patients with other neurological diseases as controls (OND, n=31). All had AD biomarker measurements available. Enzyme‐linked immunosorbent assay (ELISA) or single molecule array (Simoa) were used to quantify the biomarker levels of NPTX2, NPTXR, NfL and GFAP in CSF, as well as NfL and GFAP in serum. Statistical analyses were performed using age and sex corrected analysis of covariance (ANCOVA), generalized linear mixed models (GLMM), ROC analyses, and Spearman correlation coefficients (ρ).ResultDifferences in protein levels across diagnostic groups were found for NPTX2, NPTXR, NfL (serum and CSF), and GFAP (serum and CSF, figure 1). GLMM predicted NPTX2 and total tau (t‐tau) to best distinguish FTD from OND (p<0.002), while serum GFAP and t‐tau best distinguished FTD from AD (p<0.05). A combined panel of NPTX2, serum GFAP and t‐tau differentiated FTD from the other groups with good accuracy (table 1). Serum GFAP and NPTX2 correlated negatively in FTD (ρ =‐0.40, p<0.05) and AD (ρ=‐0.31, p<0.05), but positively in DLB (ρ =0.87, p<0.01), while no significant correlation was found in OND and PD (figure 2).ConclusionWe show the potential use of a novel biomarker panel including CSF t‐tau, CSF NPTX2 and serum GFAP, for differential diagnosis of FTD from other dementias. The included biomarkers reflect different neurological pathways, i.e. neurodegeneration, synaptic and astrocytic function, respectively. This suggests a unique interplay of these pathways in FTD and demonstrates the added value of the novel biomarkers to the established classical AD biomarkers. To assess the diagnostic potential of this novel panel, validation in a larger cohort should be performed.