%0 Journal Article
%T Association of Plasma Amyloid, P-Tau, GFAP, and NfL With CSF, Clinical, and Cognitive Features in Patients With Dementia With Lewy Bodies.
%A Bolsewig K
%A van Unnik AAJM
%A Blujdea ER
%A Gonzalez MC
%A Ashton NJ
%A Aarsland D
%A Zetterberg H
%A Padovani A
%A Bonanni L
%A Mollenhauer B
%A Schade S
%A Vandenberghe R
%A Poesen K
%A Kramberger MG
%A Paquet C
%A Bousiges O
%A Cretin B
%A Willemse EAJ
%A Teunissen CE
%A Lemstra AW
%A  
%J Neurology
%V 102
%N 12
%D 2024 Jun 25
%M 38830138
%F 11.8
%R 10.1212/WNL.0000000000209418
%X <B>OBJECTIVE: </B>Plasma β-amyloid-1-42/1-40 (Aβ42/40), phosphorylated-tau (P-tau), glial fibrillary acidic protein (GFAP), and neurofilament light (NfL) have been widely examined in Alzheimer disease (AD), but little is known about their reflection of copathologies, clinical importance, and predictive value in dementia with Lewy bodies (DLB). We aimed to evaluate associations of these biomarkers with CSF amyloid, cognition, and core features in DLB.<BR><B>METHODS: </B>This cross-sectional multicenter cohort study with prospective component included individuals with DLB, AD, and healthy controls (HCs), recruited from 2002 to 2020 with an annual follow-up of up to 5 years, from the European-Dementia With Lewy Bodies consortium. Plasma biomarkers were measured by single-molecule array (Neurology 4-Plex E kit). Amyloid status was determined by CSF Aβ42 concentrations, and cognition was assessed by Mini-Mental State Examination (MMSE). Biomarker differences across groups, associations with amyloid status, and clinical core features were assessed by analysis of covariance. Associations with cognitive impairment and decline were assessed by linear regression and linear mixed-effects models.<BR><B>RESULTS: </B>In our cohort consisting of 562 individuals (HC n = 89, DLB n = 342, AD n = 131; 250 women [44.5%], mean [SD] age of 71 [8] years), sex distribution did not differ between groups. Patients with DLB were significantly older, and had less years of education and worse baseline cognition than HC, but not AD. DLB participants stratified for amyloid status differed significantly in plasma Aβ42/40 ratio (decreased in amyloid abnormal: β = -0.008, 95% CI -0.016 to -0.0003, p = 0.01) and P-tau (increased in amyloid abnormal, P-tau181: β = 0.246, 95% CI 0.011-0.481; P-tau231: β = 0.227, 95% CI 0.035-0.419, both p < 0.05), but not in GFAP (β = 0.068, 95% CI -0.018 to 0.153, p = 0.119), and NfL (β = 0.004, 95% CI -0.087 to 0.096, p = 0.923) concentrations. Higher baseline GFAP, NfL, and P-tau concentrations were associated with lower MMSE scores in DLB, and GFAP and NfL were associated with a faster cognitive decline (GFAP: annual change of -2.11 MMSE points, 95% CI -2.88 to -1.35 MMSE points, p < 0.001; NfL: annual change of -2.13 MMSE points, 95% CI -2.97 to -1.29 MMSE points, p < 0.001). DLB participants with parkinsonism had higher concentrations of NfL (β = 0.08, 95% CI 0.02-0.14, p = 0.006) than those without.<BR><B>CONCLUSIONS: </B>Our study suggests a possible utility of plasma Aβ42/40, P-tau181, and P-tau231 as a noninvasive biomarkers to assess amyloid copathology in DLB, and plasma GFAP and NfL as monitoring biomarkers for cognitive symptoms in DLB.